# Privacy Policy
Source: https://docs.boltz.bio/privacy-policy/privacy-policy
Boltz Privacy Notice - How we collect, use and protect your information
**Last Updated:** January 6, 2026
This Privacy Notice describes how Boltz PBC (“**we**”, “**us**,” “**our**”) collects, uses and discloses information about individuals who use our website ([https://boltz.bio](https://boltz.bio)), applications, services, tools and features, or otherwise interact with us (collectively, the “**Services**”). For the purposes of this Privacy Notice, “**you**” and “**your**” means you as the user of the Services, whether you are a customer, website visitor, representative of a company with whom we do business, or another individual whose information we have collected pursuant to this Privacy Notice.
Please read this Privacy Notice carefully. By using any of the Services, you agree to the collection, use, and disclosure of your information as described in this Privacy Notice. If you do not agree to this Privacy Notice, please do not use or access the Services.
***
## Changes to This Privacy Notice
We may modify this Privacy Notice from time to time, in which case we will update the “Last Updated” date at the top of this Privacy Notice. If we make material changes to the way in which we use or disclose information we collect, we will use reasonable efforts to notify you (such as by emailing you at the last email address you provided us, by posting notice of such changes on the Services, or by other means consistent with applicable law) and will take additional steps as required by applicable law.
If you do not agree to any updates to this Privacy Notice, please do not continue using or accessing the Services.
***
## Collection and Use of Your Information
When you use or access the Services, we collect certain categories of information about you from a variety of sources.
### Information You Provide to Us
Some features of the Services may require you to directly provide us with certain information about yourself. You may elect not to provide this information, but doing so may prevent you from using or accessing these features. Information that you directly submit through our Services includes:
| Information Type | What We Collect | How We Use It |
| ------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Basic contact details** | Name, address, phone number, and email | To create and maintain your account, provide the Services, and communicate with you (including to tell you about products or services that may be of interest to you) |
| **Account information** | Username, password, security questions you select and the answers you provide | To provide the Services and to maintain and secure your account with us. |
| **Payment information** | Bank account, credit or debit card information, and billing address (collected via third-party payment processor) | To process your payment and provide the Services |
| **Any other information you choose to include in communications with us** | Any information you include, for example, when sending a message through the Services | To respond to your inquiries and provide support |
If you choose to register an account, you are responsible for keeping your account credentials safe. We recommend you do not share your access details with anyone else. If you believe your account has been compromised, please contact us immediately.
### Information We Collect Automatically
We also automatically collect certain information about your interaction with the Services ("**Usage Data**"). To do this, we may use cookies, web beacons and/or other geolocation tracking technologies ("**Tracking Technologies**"). Usage Data includes:
| Data Type | Examples |
| ------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------- |
| **Device information** | Device type, operating system, unique device identifier, and internet protocol (IP) address |
| **Location information** | Approximate location / precise geolocation, if you choose to provide it |
| **Other information regarding your interaction with the Services** | Browser type, log data, date and time stamps, clickstream data, interactions with marketing emails, and ad impressions |
We use Usage Data to tailor features and content to you, market to you, provide you with offers or promotions, run analytics and better understand user interaction with the Services. For more information on how we use Tracking Technologies and your choices, see the section below, [Cookies and Other Tracking Technologies](#cookies-and-other-tracking-technologies).
Any information we receive from outside sources will be treated in accordance with this Privacy Notice. We are not responsible for the accuracy of the information provided to us by third parties and are not responsible for any third party’s policies or practices. For more information, see the section below, [Third Party Websites and Links](#third-party-websites-and-links).
### Additional Uses
In addition to the specific uses described above, we may use any of the above information to provide you with the Services and to maintain our business relationship, including by enhancing the safety and security of our Services (e.g., troubleshooting, data analysis, testing, system maintenance, and reporting), providing customer support, sending service and other non-marketing communications, monitoring and analyzing trends, conducting internal research and development, complying with applicable legal obligations, enforcing any applicable terms of service, and protecting the Services, our rights, and the rights of our employees, users or other individuals.
### Deidentified Information
Finally, we may deidentify or anonymize your information such that it cannot reasonably be used to infer information about you or otherwise be linked to you (“**deidentified information**”) (or we may collect information that has already been deidentified/anonymized), and we may use such deidentified information for any purpose. To the extent we possess or process any deidentified information, we will maintain and use such information in deidentified/anonymized form and not attempt to re-identify the information, except solely for the purpose of determining whether our deidentification/anonymization process satisfies legal requirements.
***
## Cookies and Other Tracking Technologies
Most browsers accept cookies automatically, but you may be able to control the way in which your devices permit the use of Tracking Technologies. If you so choose, you may block or delete our cookies from your browser; however, blocking or deleting cookies may cause some of the Services, including certain features and general functionality, to work incorrectly. If you have questions regarding the specific information about you that we process or retain, as well as your choices regarding our collection and use practices, please contact us using the information listed below.
Your browser settings may allow you to transmit a “do not track” signal, “opt-out preference” signal, or other mechanism for exercising your choice regarding the collection of your information when you visit various websites. Like many websites, our website is not designed to respond to such signals, and we do not use or disclose your information in any way that would legally require us to recognize opt-out preference signals.
***
## Disclosure of Your Information
We may disclose your information to third parties for legitimate purposes subject to this Privacy Notice, including the following categories of third parties:
* Our affiliates or others within our corporate group.
* Vendors or other service providers who help us provide the Services, including for system administration, cloud storage, security, customer relationship management, marketing communications, web analytics, payment networks, and payment processing.
* Third parties for marketing purposes.
* Third parties to whom you request or direct us to disclose information, such as through your use of social media widgets or login integration.
* Professional advisors, such as auditors, law firms, or accounting firms.
* Third parties in connection with or anticipation of an asset sale, merger, or other business transaction, including in the context of a bankruptcy.
We may also disclose your information as needed to comply with applicable law or any obligations thereunder or to cooperate with law enforcement, judicial orders, and regulatory inquiries, to enforce any applicable terms of service, and to ensure the safety and security of our business, employees, and users.
***
## Third Party Websites and Links
We may provide links to third-party websites or platforms. If you follow links to sites or platforms that we do not control and are not affiliated with us, you should review the applicable privacy notice, policies and other terms. We are not responsible for the privacy or security of, or information found on, these sites or platforms. Information you provide on public or semi-public venues, such as third-party social networking platforms, may also be viewable by other users of the Services and/or users of those third-party platforms without limitation as to its use. Our inclusion of such links does not, by itself, imply any endorsement of the content on such platforms or of their owners or operators.
***
## Children's Privacy
Children under the age of 18 are not permitted to use the Services, and we do not seek or knowingly collect any personal information about children under 18 years of age.
If we become aware that we have unknowingly collected information about a child under 18 years of age, we will make commercially reasonable efforts to delete such information. If you are the parent or guardian of a child under 18 years of age who has provided us with their personal information, you may contact us using the below information to request that it be deleted.
***
## Data Security and Retention
Despite our reasonable efforts to protect your information, no security measures are impenetrable, and we cannot guarantee “perfect security.” Any information you send to us electronically, while using the Services or otherwise interacting with us, may not be secure while in transit. We recommend that you do not use unsecure channels to send us sensitive or confidential information.
We retain your information for as long as is reasonably necessary for the purposes specified in this Privacy Notice. When determining the length of time to retain your information, we consider various criteria, including whether we need the information to continue to provide you the Services, resolve a dispute, enforce our contractual agreements, prevent harm, promote safety, security and integrity, or protect ourselves, including our rights, property or products.
***
## Legal Bases for Use of Your Information
The legal bases for using your information as set out in this Privacy Notice are as follows:
* Where we need to perform the contract we are about to enter into or have entered into with you for the Services;
* Where it is necessary for our legitimate interests (or those of a third party) and your interests and fundamental rights do not override those interests;
* Where we need to comply with a legal or regulatory obligation;
* Where we have your consent to process your information in a certain way.
***
## Consent to Data Transfer
Your information is maintained and processed by us and our third-party service providers in the United States, and may also be maintained, processed, and stored in other jurisdictions that may have different data protection law than those in your country of residence. In the event that your information is transferred in these ways, please note that we comply with applicable legal requirements governing the transfer of information across borders, such as by using standard contractual clauses. By using the Services, you agree to and acknowledge these transfers.
***
## Your Privacy Rights
*Access, Update, Delete*
You have the right to request access to the personal information we collect from you, correct, change or update that information, or delete it. You may review, update, or delete your personal information by logging into your account. Requests to access, change, or delete your information will be handled within 30 days.
*Withdrawing Your Consent*
If we are relying on your consent to process your personal information, which may be express and/or implied consent depending on the applicable law, you have the right to withdraw your consent at any time. You can withdraw your consent at any time by contacting us by using the contact details provided below. However, please note that this will not affect the lawfulness of the processing before its withdrawal nor, when applicable law allows, will it affect the processing of your personal information conducted in reliance on lawful processing grounds other than consent.
*Opting Out of Marketing and Promotional Communications*
You may manage your receipt of marketing and non-transactional communications by clicking on the “unsubscribe” link located on the bottom of Boltz’s marketing emails or by contacting us by using the contact details provided below. If, at any time after registering, you change your mind about receiving information from us or about the use of information volunteered by you, please send us a request specifying your new choice. You will then be removed from the marketing lists. However, we may still communicate with you — for example, to send you service-related messages that are necessary for the administration and use of your account, to respond to service requests, or for other non-marketing purposes.
*Additional Legal Rights*
In accordance with the laws of certain jurisdictions, you may have certain rights and choices regarding the personal information we collect and maintain about you and how we communicate with you. Where the European Union General Data Protection Regulation (“GDPR”) or similar legal requirements apply to the processing of your personal information, especially when you access this website from a country in the European Economic Area (“EEA”) or in a jurisdiction with similar legal protections, you may have the following additional rights:
* The right to restrict our use of your personal information;
* The right to object to our use of, or certain types of disclosures, of your personal information;
* The right to request the transfer of your personal information we hold about you to a third party;
* The right to receive your personal information in a usable format and transmit it to a third party (also known as the right of data portability);
* The right to lodge a complaint with your local data protection authority. In the EEA, you can find those at: [https://www.edpb.europa.eu/about-edpb/about-edpb/members\_en](https://www.edpb.europa.eu/about-edpb/about-edpb/members_en). If you are located in Switzerland, the contact details for the data protection authorities are available here: [https://www.edoeb.admin.ch/edoeb/en/home.html](https://www.edoeb.admin.ch/edoeb/en/home.html). If you are located in the United Kingdom, the contact details for the Information Commissioner’s Officer are available here: [https://ico.org.uk/make-a-complaint/](https://ico.org.uk/make-a-complaint/)
If you have questions or comments about your privacy rights, you may contact us using the contact information below. We will consider all requests and provide our response within the time period required by applicable law. Please note, however, that certain information may be exempt from such requests in some circumstances, which may include if we need to keep processing your information for our legitimate interests or to comply with a legal obligation. We may request you provide us with information necessary to confirm your identity before responding to your request.
***
## How to Contact Us
Should you have any questions about our privacy practices or this Privacy Notice, please contact us:
**Boltz PBC**\
857 Beacon St, Apt. 63\
Boston, Massachusetts, 02215
**Email:** [contact@boltz.bio](mailto:contact@boltz.bio)
# Boltz Terms of Service
Source: https://docs.boltz.bio/terms-of-service/terms-of-service
Boltz Lab Terms of Service Agreement
**Last Updated**: May 6, 2026
These Terms of Service (this “**Agreement**”) are between you (“**Customer**”) and Boltz PBC (“**Boltz**”) and is effective upon the earlier of when Customer enters into an Order Form with Boltz or when Customer begins accessing or using the Services. Boltz and Customer may be referred to herein collectively as the “**Parties**” or individually as a “**Party**.”
BY ACCESSING OR USING THE SERVICES, YOU ARE EXPRESSLY ACCEPTING AND AGREEING TO THIS AGREEMENT. IF YOU ARE AN INDIVIDUAL AGREEING TO THE TERMS OF THIS AGREEMENT ON BEHALF OF AN ENTITY, SUCH AS YOUR EMPLOYER, YOU REPRESENT THAT YOU HAVE THE LEGAL AUTHORITY TO BIND THAT ENTITY. IF YOU DO NOT HAVE SUCH AUTHORITY, OR IF YOU DO NOT AGREE WITH THIS AGREEMENT, YOU MAY NOT USE THE SERVICES EITHER YOURSELF OR ON BEHALF OF THE ENTITY. BOLTZ MAY MODIFY THIS AGREEMENT FROM TIME TO TIME WITH NOTICE TO CUSTOMER. TOGETHER WITH SUCH MODIFIED TERMS, BOLTZ WILL IDENTIFY THE EFFECTIVE DATE OF THE MODIFICATIONS BY INDICATING WHEN THIS AGREEMENT WAS LAST UPDATED.
***
## 1. Definitions
Capitalized terms shall have the meaning assigned to them in this Agreement.
“**Authorized User(s)**” means individuals authorized by Customer to access and use the Services on Customer’s behalf.
“**Customer Materials**” means any data, information or other material provided, uploaded, or submitted by Customer or its Authorized Users in the course of using the Services. For avoidance of doubt, Customer Materials does not include Usage Data.
“**Documentation**” means the Services documentation, manuals, and guides made available by Boltz to Customer and updated from time to time.
“**Order Form**” means the digital or physical ordering document identifying the products and related Fees as applicable for Customer’s authorized purchases from Boltz. For clarify “Order Form” includes any Services and Fees descriptions selected by Customer for Services when creating a Customer Account or later selected by Customer through Customer’s Account. Order Forms shall be deemed incorporated herein by reference.
“**Services**” means Boltz’s software-as-a-service platform, APIs, AI models, and related technology and software, including any updates thereto, ordered by Customer and provided by Boltz pursuant to an applicable Order Form.
“**Usage Data**” means data and information related to Customer’s or its Authorized Users’ use of the Services or Customer Materials which Boltz uses solely in a deidentified form to derive data or insights. For avoidance of doubt, Usage Data does not include Customer Materials that have not been anonymized and Boltz will not use any Customer Materials to train its models without Customer’s prior authorization.
***
## 2. Scope of Services and Restrictions
### 2.1 Right to Use
Subject to Customer’s compliance with the terms and conditions of this Agreement and the applicable Order Form, Boltz hereby grants Customer a limited, non-exclusive, non-transferable, non-sublicensable right to permit its Authorized Users to access and use the Services and Documentation during the applicable Order Form term for Customer’s internal business purposes.
### 2.2 Use Restrictions
Customer shall not use the Services for any purposes beyond the scope of the rights granted in this Agreement. Without limiting the foregoing and except as expressly set forth in this Agreement, Customer shall not and shall not permit any third party to: (a) modify, adapt, alter, translate, or create derivative works of the Services; (b) reverse-engineer, decompile, disassemble, or attempt to derive the source code for the Services, in whole or in part, except to the extent that such activities are permitted under applicable law; (c) distribute, license, sublicense, lease, rent, loan, or otherwise transfer the Services to any third party; (d) remove, alter, or obscure in any way any notices (including copyright, patent, and trademark notices and symbols) of Boltz or its licensors or suppliers contained on or within any copies of the Services or Documentation; (e) use the Services for the purpose of creating a product or service competitive with the Services; (f) attempt to gain unauthorized access to, interfere with, damage or disrupt the Services, accounts registered to other users, or the computer systems or networks connected to the Services; (g) circumvent, remove, alter, deactivate, degrade or thwart any technological measure or content protections of the Services; or (h) use the Services other than as described in this Agreement or the applicable Documentation, or in violation of applicable laws. Customer shall be responsible for any violation of these terms by its Authorized Users.
### 2.3 Account Management
To use the Services, Customer will be required to create an account (“Account”). Customer agrees to provide Boltz with accurate, complete, and updated information for Customer’s Account. Customer is solely responsible for any activity on its Account and for maintaining the confidentiality and security of its password. Boltz is not liable for any acts or omissions by Customer or its Authorized Users in connection with Customer’s Account. Customer must promptly notify Boltz if it knows or has any reason to suspect that Customer’s Account or password has been stolen, misappropriated, or otherwise compromised, or in case of any actual or suspected unauthorized use of Customer’s Account. Customer agrees not to create any Account if Boltz has previously removed Customer’s Account, or Boltz previously banned Customer from any of the Services, unless Boltz provides written consent otherwise. Boltz may collect, process, and use the Account information you provide to Boltz in accordance with Boltz’s Privacy Notice, and by accessing and using the Services, Customer agrees to be bound by the terms of the Privacy Notice (as may be updated from time to time). Boltz’s Privacy Notice is available at: [https://boltz.bio/privacy](https://boltz.bio/privacy).
### 2.4 Publications
Customer acknowledges and agrees that, to the extent that Customer publishes any paper or other content using the Services, that Customer will acknowledge Boltz and cite Boltz’s relevant publications.
### 2.5 Suspension
Boltz may suspend Customer’s use of the Services if Boltz reasonably believes that the use of the Services by Customer poses a security risk to the Services or Boltz’s other clients or users, or Customer is otherwise is in violation of the terms of this Agreement. Boltz will give Customer notice before suspending Customer’s use of the Services, unless a security risk is immediate or notice is prohibited by law, in which case Boltz will provide Customer with notice as soon permitted. Customer’s access to the Services will be reinstated promptly once the issue causing the suspension has been resolved.
### 2.6 Boltz API
Boltz may provide functionality that permits Customer to access certain Boltz AI models via the Boltz API. In the event Customer utilizes the API to access Boltz AI models, Customer agrees to be bound by the terms and conditions of the Boltz API Supplemental Terms available at: [API Supplemental Terms](https://api.boltz.bio/docs/legal/api-supplemental-terms).
### 2.7 Trial Services
If Customer is accessing or using the Services on a no-fee, evaluation, proof-of-concept, or trial basis (“Trial Services”), Customer acknowledges and agrees that Boltz may modify or terminate the Trial Services at any time in its sole discretion, and the Trial Services may be subject to certain limitations and restrictions, or additional terms, as specified by Boltz. Customer is only permitted to use Trial Services during the limited period as designated by Boltz (or, if not designated, until terminated in accordance with this Section). To the maximum extent permitted by law, Boltz disclaims all obligations and liabilities with respect to Trial Services, including but not limited to any warranty or indemnity obligations.
TRIAL SERVICES ARE PROVIDED ON AN “AS-IS” BASIS. NOTWITHSTANDING SECTION 10, OR ANY OTHER TERM TO THE CONTRARY IN THIS AGREEMENT, BOLTZ’S TOTAL LIABILITY TO CUSTOMER ARISING FROM THE TRIAL SERVICES WILL NOT EXCEED ONE THOUSAND DOLLARS (\$1,000).
***
## 3. Fees
### 3.1 Fees
Customer shall pay to Boltz the fees as set forth in each applicable Order Form (“Fees”) and will provide accurate and current billing contact information. Customer acknowledges and agrees that, except as otherwise expressly set forth in this Agreement, Fees are non-refundable. In the event Customer exceeds any usage amounts in the applicable Order Form, Customer shall be responsible for the Fees associated with such excess use.
### 3.2 Invoicing and Payment
Boltz will invoice Customer as otherwise specified in the applicable Order Form. Unless otherwise set forth in an applicable Order Form, Customer shall pay all invoices within thirty (30) days of the invoice date. If Customer provides credit card or debit card information for payment purposes, Customer authorizes Boltz, and its third-party payment processor (Stripe), to charge Customer’s credit card, debit card, or other means of payment for all Fees, including Fees associated with excess use. Customer acknowledges and agrees that payment by credit card or debit card may be subject to separate terms between Customer and Stripe. Boltz will not be liable for the acts or omissions of Stripe. Customer shall make all payments hereunder in US dollars. If any invoiced amount is not received by Boltz by the due date, then without limiting Boltz's rights or remedies those charges may accrue late interest at the rate of 1.5% of the outstanding balance per month, or the maximum rate permitted by law, whichever is lower.
### 3.3 Taxes
All Fees and other amounts payable by Customer under this Agreement are exclusive of taxes and similar assessments. Customer is responsible for all sales, use, and excise taxes, and any other similar taxes, duties, and charges of any kind imposed by any federal, state, or local governmental or regulatory authority on any amounts payable by Customer hereunder, other than any taxes imposed on Boltz's income.
***
## 4. Intellectual Property Rights
### 4.1 Ownership of the Services
The Services, including their “look and feel” (e.g., text, graphics, images, logos), proprietary content, information and other materials, are protected under copyright, trademark, and other intellectual property laws. Customer agrees that Boltz and/or its licensors own all right, title, and interest in and to the Services (including any and all intellectual property rights therein) and Customer agrees not to take any actions inconsistent with such ownership interests. Boltz and its licensors reserve all rights in connection with the Services and its content (other than Customer Materials and Outputs), including, without limitation, the exclusive right to create derivative works.
### 4.2 Customer Materials
Customer Materials will be deemed Confidential Information of Customer and as between Customer and Boltz, Customer owns and retains all right, title and interest in and to all Customer Materials, including all intellectual property rights therein. Customer represents and warrants that Boltz’s use of the Customer Materials in accordance with this Agreement and in the form provided to Boltz by Customer will not violate any applicable laws or regulations, infringe or violate any intellectual property or other rights of any third party, or cause a breach of any agreement or obligations between Customer and any third party. Customer further represents and warrants that it has all rights and permissions required to submit Input to the Services.
### 4.3 Additional Input and Output Specific Terms
The Services will generate outputs (each, an “Output”) in response to Customer Materials uploaded to the Services (collectively, “Input”). The Input and Output will be deemed Confidential Information of Customer. As between the Parties, to the extent permitted by applicable law and subject to Section 4.1: (a) Customer owns all Input provided by Customer; and (b) subject to terms and conditions of this Agreement and Customer’s compliance therewith, Customer may use Outputs for its legally permitted business purposes.
Customer may not use Output to develop artificial intelligence or machine learning models that compete with Boltz.
### 4.4 Usage Data
Boltz owns all Usage Data. Boltz may freely use Usage Data in connection with improving, testing, and operating the Services, provided that Usage Data is in aggregate and/or deidentified form and cannot be linked specifically to Customer.
### 4.5 Feedback
From time-to-time Customer or its Authorized Users may provide Boltz with suggestions, comments, feedback or the like regarding the Services (collectively, “Feedback”). Customer hereby grants Boltz a perpetual, irrevocable, royalty-free and fully-paid up license to use and exploit all Feedback in connection with Boltz’s business purposes, including, without limitation, the testing, development, maintenance and improvement of the Services.
### 4.6 Third-Party Materials
Certain Services may display, include or make available content, data, information, applications, services, products, or materials from third parties (“Third-Party Materials”). By using the Services, you acknowledge and agree that Boltz is not responsible for examining or evaluating the content, accuracy, completeness, availability, timeliness, validity, copyright compliance, legality, decency, quality or any other aspect of such Third-Party Materials. Boltz does not warrant and does not assume and will not have any liability or responsibility to Customer or any other person for any Third-Party Materials. Any part of the Services that contains or utilizes open-source software is distributed and made available under the terms of the open-source license agreements referenced in the applicable distribution or the applicable help, notices, about or source files or Documentation. Copyrights and other proprietary rights to the open-source software are held by the copyright holders identified in the applicable distribution or the applicable help, notices, about, source files, or Documentation. The Services shall not include any code licensed under any “viral” or “copyleft” license.
***
## 5. Term and Termination
### 5.1 Term
This Agreement begins on Effective Date and will remain in effect until it is terminated in accordance with the terms herein (the “Term”).
### 5.2 Termination
This Agreement and the Order Forms, may be terminated by either Party: (a) if there are no currently active or in effect Order Forms, upon thirty (30) days’ written notice to the other Party; (b) effective on written notice to the other Party, if the other Party materially breaches this Agreement, and such breach (i) is incapable of cure, or (ii) being capable of cure, remains uncured thirty (30) days after the non-breaching Party provides the breaching Party with written notice of such breach; or (c) upon written notice to the other Party if the other Party (i) has made or attempted to make any assignment for the benefit of its creditors or any compositions with creditors, (ii) has any action or proceedings under any bankruptcy or insolvency laws taken by or against it which have not been dismissed within sixty (60) days, (iii) has effected a compulsory or voluntary liquidation or dissolution, or (iv) has undergone the occurrence of any event analogous to any of the foregoing under the law of any jurisdiction.
### 5.3 Effect of Termination
Upon any expiration or termination of this Agreement, Customer shall (a) immediately cease use of the Services, and (b) return all Boltz Confidential Information and other materials and information provided by Boltz. Any termination or expiration shall not relieve Customer of its obligation to pay all Fees accrued prior to termination. This Section 5.3 and Sections 1, 4, 6, 8-12 will survive any termination or expiration of this Agreement.
***
## 6. Confidentiality
### 6.1 Confidentiality
As used herein, “Confidential Information” means any information that one Party (the “Disclosing Party”) provides to the other Party (the “Receiving Party”) in connection with this Agreement, whether orally or in writing, that is designated as confidential or that reasonably should be considered to be confidential given the nature of the information and/or the circumstances of disclosure. For clarity, the Services and Documentation will be deemed Confidential Information of Boltz and Customer Materials, including Inputs, and Outputs will be deemed Confidential Information of Customer. The Receiving Party will not use or disclose any Confidential Information of the Disclosing Party except as expressly permitted herein and as necessary to perform its obligations or exercise its rights under this Agreement. The Receiving Party shall exercise the same degree of care to prevent unauthorized use or disclosure of Discloser's Confidential Information to others as it takes to preserve and safeguard its own information of like importance, but in no event less than reasonable care. The Receiving Party may disclose Confidential Information of the Disclosing Party only to those of its employees, contractors, agents and advisors (collectively, “Representatives”) who have a bona fide need to know such Confidential Information to perform under this Agreement, who have been informed of its confidential nature, and who are bound by confidentiality obligations at least as protective as those set forth in this Agreement. The Receiving Party shall be responsible for any breach of this Section 6 caused by its Representatives.
### 6.2 Exclusions
Confidential Information will not include any information that: (a) is or becomes generally known to the public through no fault or breach of this Agreement by the Receiving Party or its Representatives; (b) is rightfully known by the Receiving Party at the time of disclosure without an obligation of confidentiality; (c) is independently developed by the Receiving Party without access to or use of any Confidential Information of the Disclosing Party that can be evidenced in writing; or (d) is rightfully obtained by the Receiving Party from a third party without restriction on use or disclosure.
### 6.3 Compelled Disclosures
If the Receiving Party is compelled by a court or other competent authority or applicable law to disclose Confidential Information of the Disclosing Party, it shall, to the extent permitted by applicable law, give the Disclosing Party prompt written notice and shall provide the Disclosing Party with reasonable cooperation so that the Disclosing Party may take steps to oppose such disclosure or obtain a protective order. If after providing such notice and assistance the Receiving Party remains required to disclose the Disclosing Party’s Confidential Information, the Receiving Party shall use all reasonable efforts to limit the disclosure and maintain the confidentiality of such Confidential Information.
### 6.4 Equitable Relief
Each Party acknowledges and agrees that a breach or threatened breach by such Party of its obligations in this Section 6 or, in the case of Customer, its breach of the restrictions in Section 2.2, may cause the other Party irreparable harm for which monetary damages may not be an adequate remedy and agrees that, in the event of such breach or threatened breach, the other Party will be entitled to equitable relief, including a restraining order, an injunction, specific performance, and any other relief that may be available from any court, without any requirement to post a bond or other security, or to prove actual damages or that monetary damages are not an adequate remedy. Such remedies are not exclusive and are in addition to all other remedies that may be available at law, in equity, or otherwise.
***
## 7. Data Security
Boltz will maintain reasonable administrative, physical, and technical security measures consistent with applicable law that are intended to protect against the loss, misuse, unauthorized access, alteration or disclosure of Customer Materials or the Services.
***
## 8. Disclaimers
### 8.1 General
THE SERVICES ARE PROVIDED "AS IS" AND BOLTZ HEREBY DISCLAIMS ALL WARRANTIES, WHETHER EXPRESS, IMPLIED, STATUTORY, OR OTHERWISE. BOLTZ SPECIFICALLY DISCLAIMS ALL IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, AND NON-INFRINGEMENT, AND ALL WARRANTIES ARISING FROM COURSE OF DEALING, USAGE, OR TRADE PRACTICE. BOLTZ MAKES NO WARRANTY OF ANY KIND THAT THE SERVICES OR RESULTS OF THE USE THEREOF, WILL MEET CUSTOMER'S OR ANY OTHER PERSON'S REQUIREMENTS, OPERATE WITHOUT INTERRUPTION, ACHIEVE ANY INTENDED RESULT, BE COMPATIBLE OR WORK WITH ANY SOFTWARE, SYSTEM, OR OTHER SERVICES, OR BE SECURE, ACCURATE, COMPLETE, FREE OF HARMFUL CODE, OR ERROR FREE.
### 8.2 Similarity, Accuracy and Appropriateness of Output
Customer acknowledges and understands that the Services utilizes artificial intelligence models. Due to their nature, Output may not be unique and the Services may generate the same or similar output for Boltz or a third party, including other Boltz clients or users.
GIVEN THE PROBABILISTIC NATURE OF ARTIFICIAL INTELLIGENCE, THE SERVICES MAY IN SOME SITUATIONS PRODUCE OUTPUT THAT IS INACCURATE, INCORRECT, OR OTHERWISE UNDESIRABLE. THE ACCURACY, QUALITY, AND COMPLIANCE WITH APPLICABLE LAW OF THE OUTPUT IS DEPENDENT UPON AND COMMENSURATE WITH THAT OF THE INPUT PROVIDED AND CUSTOMER’S COMPLIANCE WITH THIS AGREEMENT, AND NOTWITHSTANDING ANYTHING ELSE SET OUT HEREIN, BOLTZ WILL NOT HAVE ANY LIABILITY OR RESPONSIBILITY TO CUSTOMER OR ANY OTHER PERSON OR ENTITY FOR ANY LOSS OR DAMAGES RELATING TO OR ARISING FROM INPUT, THE OUTPUT, OR THEIR USE.
Customer shall evaluate the content, nature, and accuracy of any Output as appropriate for the applicable use case, including by using human review of the Output.
***
## 9. Indemnification
### 9.1 Indemnification by Boltz
Boltz will defend, indemnify and hold Customer harmless from and against any and all losses, damages, liabilities, costs, and expenses, including reasonable attorneys’ fees (“Losses”) incurred by Customer resulting from a claim, suit or proceeding brought by a third-party (“Claims”) to the extent such Losses arise or result from allegations that the Services infringe or misappropriates such third party’s intellectual property rights. If Boltz reasonably believes the Services (or any component thereof) could infringe any third party’s intellectual property rights, Boltz may, at its sole option and expense use commercially reasonable efforts to: (a) modify or replace the Services, or any component or part thereof, to make it non-infringing; or (b) procure the right for Customer to continue using the Services. If Boltz determines that neither alternative is commercially practicable, Boltz may terminate this Agreement, in its entirety or with respect to the affected component, by providing written notice to Customer. In the event of any such termination, Boltz will refund to Customer a pro-rata portion of the Fees that have been paid for the unexpired portion of the then current Term.
THE RIGHTS AND REMEDIES SET FORTH IN THIS SECTION 9.1 WILL CONSTITUTE CUSTOMER’S SOLE AND EXCLUSIVE REMEDY FOR ANY INFRINGEMENT OR MISAPPROPRIATION OF INTELLECTUAL PROPERTY RIGHTS IN CONNECTION WITH THE SERVICES.
### 9.2 Exclusions
Boltz’s obligations under Section 9.1 will not apply if the underlying Claim arises from or as a result of: (a) Customer’s breach of this Agreement, negligence, willful misconduct or fraud or failure to use the Services in accordance with the Documentation; (b) any Customer Materials, Input, or Output; (c) Customer’s failure to use any enhancements, modifications, or updates to the Services that have been provided by Boltz; (d) modifications to or configuration of the Services by anyone other than Boltz or otherwise specifically authorized by Boltz in writing (email being sufficient); or (e) combinations of the Services with software, data or materials not provided by Boltz or not in accordance with the Documentation.
### 9.3 Indemnification by Customer
Customer will defend, indemnify and hold Boltz and its affiliates, officers, directors, agents, and employees (“Boltz Indemnitees”) harmless from and against any and all Losses incurred by Boltz Indemnitees arising or resulting from Claims arising or relating from: (a) allegations that the Customer Materials, Input, or their use by Boltz in accordance with this Agreement infringes, misappropriates or violates a third party’s intellectual property rights, or rights of publicity or privacy, or result in the violation of any applicable law or regulation; (b) Customer’s or an Authorized User’s use of the Services to the extent such use was not in accordance with this Agreement or Documentation; (c) Customer’s use of the Output, and (d) the manufacture, sale, distribution, or marketing of any Customer’s products or services.
### 9.4 Procedure
The Party seeking defense and indemnity (the “Indemnified Party”) will provide the other Party (the “Indemnifying Party”) with prompt written notice of such Claim. The Indemnifying Party will have the right to defend or settle such Claim, provided it will not make any settlement of a Claim that results in any liability or imposes any obligation on the Indemnified Party without the prior written consent of such Indemnified Party, which will not be unreasonably withheld. The Indemnified Party will reasonably cooperate with the Indemnifying Party, at the Indemnifying Party’s expense, in the defense and settlement of such Claim. The Indemnified Party may participate in the defense of any Claim at its own expense. An Indemnified Party’s failure to perform obligations under this Section 9.4 will not relieve the Indemnifying Party of its obligations under Section 9 except to the extent that Indemnifying Party is materially prejudiced as a result of such failure.
***
## 10. Limitation of Liability
TO THE EXTENT PERMITTED BY APPLICABLE LAW, NEITHER PARTY WILL BE LIABLE TO THE OTHER PARTY FOR ANY INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, PUNITIVE OR CONSEQUENTIAL DAMAGES, OR ANY LOSS OF INCOME, DATA, PROFITS, REVENUE OR BUSINESS INTERRUPTION, OR THE COST OF COVER OR SUBSTITUTE SERVICES, ARISING OUT OF OR IN CONNECTION WITH THIS AGREEMENT. TO THE EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT WILL BOLTZ’S TOTAL CUMULATIVE AGGREGATE LIABILITY ARISING OUT OF OR RELATING TO THIS AGREEMENT EXCEED THE CUMULATIVE FEES ACTUALLY PAID OR PAYABLE BY CUSTOMER TO BOLTZ IN THE TWELVE (12) MONTH PERIOD PRECEDING THE EVENT GIVING RISE TO THE CLAIM. THE LIMITATIONS AND EXCLUSIONS IN THIS SECTION WILL APPLY WHETHER SUCH LIABILITY ARISES FROM ANY CLAIM BASED ON CONTRACT, WARRANTY, TORT (INCLUDING NEGLIGENCE), STRICT LIABILITY OR OTHERWISE, AND WHETHER OR NOT BOLTZ WAS ADVISED OF THE POSSIBILITY OF SUCH LOSS OR DAMAGE.
***
## 11. Export Restrictions
Customer understands that the Services are subject to United States export controls administered by the United States Department of Commerce and the United States Department of Treasury Office of Foreign Assets Control. Customer acknowledges and agrees that the Services may not be used, transferred, or otherwise exported or re-exported to countries as to which the United States, maintains an embargo (collectively, “Embargoed Countries”), or to or by a national or resident thereof, or any person or entity on the U.S. Department of Treasury’s List of Specially Designated Nationals or the U.S. Department of Commerce’s Entity List, Denied Persons List, or Unverified List, or the U.S. Department of State’s Nonproliferation Sanctions list (collectively, “Designated Nationals”). The lists of Embargoed Countries and Designated Nationals are subject to change without notice. By using the Services, Customer represents and warrants that Customer is not located in, under the control of, or a national or resident of an Embargoed Country or Designated National. Customer agrees to comply with all United States export laws and assumes sole responsibility for obtaining United States government export licenses to export or re-export as may be required. Customer will defend, indemnify, and hold Boltz and its licensors harmless from and against any liabilities arising from Customer’s or any of its officers, directors, employees, agents, or representatives violation of such laws or regulations.
***
## 12. General
### 12.1 Relationship
The Parties are independent contractors. This Agreement shall not establish any relationship of partnership, joint venture, employment, franchise, or agency between the Parties.
### 12.2 Third-Party Beneficiaries
Nothing in this Agreement shall confer, or is intended to confer, on any third party any benefit or the right to enforce any term or condition contained herein.
### 12.3 Publicity
Subject to Customer’s prior authorization, Customer hereby grants Boltz the right to identify Customer as a Boltz Customer, and use Customer's name, mark and logo on Boltz’s website and marketing materials.
### 12.4 Assignment
Neither Party may assign any of its rights or obligations under this Agreement, whether by operation of law or otherwise, without the prior written consent of the other Party (not to be unreasonably withheld), provided that a Party may assign this Agreement, without the other Party’s consent, to an affiliate or in connection with a merger, acquisition, corporate reorganization, or sale of all or substantially all of its assets.
### 12.5 Force Majeure
Except for the inability to meet financial obligations, neither Party shall be responsible for failure or delay in performance by events out of their reasonable control, including but not limited to, pandemics, Internet outage, terrorism, war, fires, earthquakes, and other disasters. The Parties agree to use their best efforts to minimize the effects of such failures or delays.
### 12.6 Notices
All notices between the Parties shall be in writing and shall be deemed to have been given if personally delivered or sent by registered or certified mail (return receipt), or by recognized courier service. Customer agrees that all notices, disclosures, and other communications that are electronically provided satisfy any legal requirement that such communications be in writing. Notices from Boltz will be considered delivered to Customer and effective as of the time it is sent to the email address Customer uses to register its Account, or from which it otherwise emails Boltz. Customer shall send notices to Boltz at: 857 Beacon St, Apt. 63, Boston, Massachusetts, 02215.
### 12.7 Entire Agreement
This Agreement, including all Order Forms, constitutes the sole and entire agreement of the Parties with respect to the subject matter of herein and supersedes all prior and contemporaneous understandings, agreements, and representations and warranties, both written and oral, with respect to such subject matter. The terms of an Order Form will prevail over the general terms herein only if such Order Form expressly references this Agreement, the intent to prevail over this Agreement, and is authorized by both Boltz and Customer. Section headings are for convenience only and shall not affect interpretation of the relevant section. No provision of any purchase order or other form employed or provided by Customer will supersede the terms and conditions of this Agreement, and any such document relating to this Agreement shall be for administrative purposes only and shall have no legal effect. If any provision of this Agreement is held invalid or unenforceable, the remainder of this Agreement will continue in full force and effect.
### 12.8 Amendments and Waiver
Customer acknowledges and agrees that Boltz has the right, in its sole discretion, to modify this Agreement from time to time, and that such modified terms will become effective upon posting. Boltz will notify Customer of modifications through Customer’s Account in the Services or email address provided to Boltz by Customer. Customer’s continued use of the Services after the effective date of the modifications will be deemed acceptance of the modified terms. No waiver will be implied from conduct or failure to enforce or exercise rights under this Agreement, nor will any waiver be effective unless in a writing signed by a duly authorized representative on behalf of the Party claimed to have waived.
### 12.9 Governing Law
This Agreement shall be governed exclusively by, and construed exclusively in accordance with, the laws of the State of Delaware, without regard to its conflict of laws provisions. Any legal suit, action, or proceeding arising out of this Agreement will be instituted exclusively in the state and federal courts of the State of Delaware, and each party irrevocably submits to the exclusive jurisdiction of such courts.
# Billing
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Billing
Manage your subscription, seats, payment methods, and invoices
The Billing page provides an overview of your available credits, payment methods, and invoice history.
## Overview
### On-Demand Billing
On-Demand Billing allows you to continue using the platform after your included compute credits are exhausted.
| Setting | Description |
| ------------ | ------------------------------------------------- |
| **Enabled** | Usage continues and is billed at on-demand rates |
| **Disabled** | Platform access pauses when credits are exhausted |
Toggle On-Demand Billing to control whether you want automatic continuation of service.
With On-Demand Billing enabled, charges can accumulate based on usage. Monitor your [Usage](/settings/usage) page regularly to track spending.
***
## Payment Methods
Add payment methods to enable billing for seats and on-demand usage.
### Adding a Payment Method
1. Click **+ Add Payment Method**
2. Enter your credit card or payment details
3. Confirm the payment method
Payment information is securely processed and stored by our payment provider.
### Managing Payment Methods
* Set a default payment method for automatic billing
* Remove outdated payment methods
* Update card details when cards expire
***
## Invoices
The Invoices section displays your billing history with the following information:
| Column | Description |
| ----------- | ----------------------------- |
| **Date** | Invoice date |
| **Status** | Payment status |
| **Amount** | Total invoice amount |
| **Invoice** | Download link for invoice PDF |
### Downloading Invoices
Click **Download** next to any paid invoice to get a PDF for your records. Invoices include:
* Itemized charges (seats, on-demand usage)
* Service period
* Payment details
* Tax information (if applicable)
***
## Pricing
For details on inference pricing and enterprise plans, please refer to our [Pricing Page](https://boltz.bio/pricing) or [contact our sales team](support@boltz.bio).
***
Billing management requires **Manage Members, Permissions, and Billing** admin access.
# General
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/General
Configure your organization name and basic settings
The General settings page allows you to configure basic information about your organization.
![Image]()
### Organization Name
Your organization name is displayed throughout the platform and helps identify your team's workspace.
### Updating Your Organization Name
1. Navigate to **Settings → General**
2. Enter your new organization name in the text field
3. Click **Save** to apply changes
Changing your organization name does not affect your data or projects.
It is purely a display name change.
***
Only users with **Admin** access to organization management can modify general settings. See [Permission Groups](/settings/permission-groups) for details.
# Members
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Members
Invite and manage team members in your organization
The Members page allows you to invite new users, manage existing team members, and assign permission groups.
## Member List
The members table displays all users in your organization with the following information:
| Column | Description |
| --------------------- | ------------------------------------------------------------------ |
| **Name** | User's full name and email address |
| **Role Badge** | Shows **Owner** for the organization owner |
| **Permission Groups** | Groups the member belongs to (e.g., Admin, All Members, Scientist) |
| **Joined At** | Date the member joined the organization |
***
## Inviting New Members
To add new team members to your organization:
1. Click the **+ Invite Members** button
2. Enter the email address of the person you want to invite
3. Select the permission groups to assign
4. Send the invitation
Invited users will receive an email with instructions to create their account and join your organization. They appear in the members list with an **Invited** badge until they accept.
***
## Managing Member Permissions
Click on the three dots that appear when hovering over any member row to open the mini-menu.
Edit and view details opens the **Member details** dialog where you can modify their permission groups.
### Permission Groups
Available groups include:
* **Admin** - Full organization management access
* **Scientist** - Project creation and management
* **Viewer** - Read-only access
* **All Members** - Base group for all users Remember to Save changes to take effect
Removing a user from permission groups may lock them out of features.
Ensure to check the Permission Groups tab to see and modify what features are applicable to each group.
***
## Searching Members
Use the search bar to quickly find members by name or email address. This is particularly useful for larger organizations.
***
## Removing Members
To remove a member from your organization:
1. Click on the three dots that appear when hovering over any member row to open the mini-menu
2. Select the option to Remove user from organization
3. Confirm the removal
Removing a member revokes their access immediately. Any predictions or work they created remains in the organization but they can no longer access it.
Invites sent to users who are yet to accept can also be revoked using this process
The **Manage Members** permission in Permission Groups controls who can invite and manage team members.
# Overview
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Overview
Manage your organization, billing, and personal preferences
The Settings area allows you to configure your Boltz Lab experience. Settings are divided into two main sections: **Organization** settings (shared across your team) and **Personal** settings (specific to your account).
## Organization Settings
Organization settings apply to your entire team and are managed by users with appropriate permissions.
Organization name and basic configuration
Invite team members and manage access
Define roles and access levels
Check credits and manage payment methods
Monitor compute usage across projects
## Personal Settings
Personal settings affect only your individual account.
Update your name, appearance preferences, and password
***
## Quick Access
Access Settings from the main navigation menu or directly at `https://app.boltz.bio/settings`.
Organization Owners and Admins have full access to all settings. Other members may have restricted access based on their Permission Group.
# Permission Groups
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Permission-Groups
Define roles and control access levels across your organization
Permission Groups allow you to define roles with specific access levels for organization management and project access. Assign members to groups to control what they can see and do.
## Default Permission Groups
Boltz provides four default permission groups that cover common organizational roles:
| Group | Admin Access | Project Access | Typical Use |
| --------------- | ---------------------------------------- | ---------------------- | -------------------------------------- |
| **Admin** | Manage Members, Permissions, and Billing | Own | Organization administrators and owners |
| **All Members** | None | Write | Standard team members |
| **Scientist** | None | Own (Write and Delete) | Project Leads and Designers |
| **Viewer** | None | Read | External collaborators, stakeholders |
***
## Editing Permissions
### Understanding Permission Levels
### Admin Access
Controls access to organization-level settings and management features.
| Level | Description |
| -------------------------------------------- | ------------------------------------------------------------------------------------- |
| **No access** | Cannot access any organization management features |
| **Manage Members** | Can invite and remove members, assign permission groups |
| **Manage Members, Permissions, and Billing** | Full organization administration including billing and permission group configuration |
### Project Access
Controls what members can do within Design Projects.
| Level | Description |
| -------------------------- | ---------------------------------------------------------------------------------- |
| **Read** | View projects, experiments, and results. Cannot create or modify. |
| **Write** | Create experiments, run virtual screens, triage molecules. Cannot delete projects. |
| **Own (write and delete)** | Full project control including creation and deletion. |
***
## Configuring Permission Groups
### Viewing Group Members
Click the **Members** button next to any group to see which users are assigned to it.
Click the **Permissions** button to modify a group's access levels:
1. **Organization management** - Select the admin access level
2. **Project access** - Select the project access level
3. Click **Save** to apply changes
Be careful when modifying or assigning the group permissions so that users only have access to appropriate content
***
## Permission Group Assignments
Members can belong to multiple permission groups. When a member belongs to multiple groups, they receive the **highest** permission level from any of their groups.
**Example:** A user in both "All Members" (Write) and "Scientist" (Own) groups will have **Own** access to projects.
### Common Configurations
* **Principal Investigators** → Admin + Scientist groups
* **Postdocs/Senior Scientists** → Scientist group
* **Graduate Students** → All Members group
* **External Collaborators** → Viewer group
* **Department Heads** → Admin group
* **Computational Chemists** → Scientist group
* **Medicinal Chemists** → All Members group
* **Management/Stakeholders** → Viewer group
* **Account Managers** → Admin group
* **Project Scientists** → Scientist group
* **Clients (read-only access)** → Viewer group
***
## Creating Custom Permission Groups
Custom permission group creation may be available.
Contact support for details.
***
## Best Practices
**Principle of least privilege**: Assign the minimum permissions necessary for each role. Start with restrictive access and expand as needed.
**Use All Members as a baseline**: Keep all users in the "All Members" group for basic platform access, then add specialized groups for elevated permissions.
**Document your permission structure**: For larger organizations, maintain documentation of which roles map to which permission groups.
***
# Profile
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Profile
Manage your personal account settings and preferences
The Profile page allows you to manage your personal account settings, including your display name, appearance preferences, and password.
## Account Information
### Email
Your email address is used for:
* Logging into the platform
* Receiving notifications and updates
* Password reset requests
Your email address cannot be changed. Contact [support@boltz.bio](mailto:support@boltz.bio) if you need further support.
### Full Name
Your display name appears throughout the platform, including:
* Member lists
* Prediction history
* Project activity logs
* Triage annotations
To update your name:
1. Edit the **Full Name** field
2. Click **Save**
***
## Appearance
Customize the platform's visual theme to match your preference. Changes apply immediately across all platform pages.
| Option | Description |
| ---------- | ------------------------------------------------------ |
| **Light** | Bright theme with white backgrounds |
| **Dark** | Dark theme with reduced eye strain |
| **System** | Automatically matches your operating system preference |
***
## Change Password
Update your account password for security.
1. Enter your **Current password**
2. Enter your **New password**
3. Click **Change Password**
### Password Security Tips
* At least 12 characters
* Mix of uppercase, lowercase, numbers, and symbols
* Avoid dictionary words and personal information
Password Managers generate and store strong, unique passwords for each service.
***
## Forgot Password?
If you've forgotten your password:
1. Log out of the platform (if logged in)
2. Click **Forgot Password** on the login page
3. Enter your email address
4. Check your email for reset instructions
5. Follow the link to create a new password
Password reset links expire after a limited time for security. Request a new link if yours has expired.
***
## Account Security
### Leaving an Organization
To leave an organization:
1. Contact your organization's Admin
2. They can remove you from the Members list
3. You'll lose access to all organization data and projects
If you are the **Owner** of an organization, you must transfer ownership before leaving. Organizations cannot exist without an owner.
***
## Data and Privacy
Your personal data is handled according to the [Privacy Policy](/legal/privacy-policy).
Organization Admins can see member names and emails in the [Members](/settings/members) page, but cannot access or modify individual profile settings.
# Usage
Source: https://docs.boltz.bio/user-guide/Settings-and-Administration/Usage
Monitor compute usage across projects and sandboxes
The Usage page provides detailed visibility into your organization's compute consumption, broken down by project and user. Any plan includes a monthly allocation of compute credits on top of which usage is billed on-demand.
## Usage Overview
The top section displays your total inference usage for the current billing period.
| Metric | Description |
| ---------------------- | -------------------------------------------------- |
| **Total compute** | Total dollar usage used this period |
| **Included remaining** | Credits remaining from your subscription allowance |
| **Billing period** | Current billing cycle dates |
### Usage Bar
The visual bar shows the proportion of **Included** vs **On-demand** usage:
* **Blue (Included)**: Usage covered by your subscription credits
* **Orange (On-demand)**: Usage beyond included credits
On-demand charges only apply if you have [On-Demand Billing](/settings/billing) enabled and have exceeded your included credits.
***
## Usage by Project
Track compute consumption for each Design Project in your organization.
| Column | Description |
| ------------- | ------------------------------------------- |
| **Project** | Name of the Design Project |
| **Type** | Project type: Small Molecule, Protein, etc. |
| **Included** | Compute costs covered by included credits |
| **On-demand** | Compute costs on-demand |
| **Total** | Combined compute cost for the project |
### Understanding Project Costs
Costs reflect the compute resources consumed by:
* Structure predictions
* Affinity calculations
* Virtual screening campaigns
* Generative molecule design
Thanks to major performance optimizations from our team, individual predictions cost only a couple of cents — significantly cheaper than alternative providers and lower than the effective cost of running and scaling the open-source models yourself.
***
## Usage by Sandbox
Track personal Sandbox usage for individual team members.
| Column | Description |
| ------------- | -------------------------------------------- |
| **Sandbox** | User name (owner of the Sandbox) |
| **Type** | Always "Sandbox" |
| **Included** | Compute costs covered by included credits |
| **On-demand** | Compute costs on-demand |
| **Total** | Combined compute cost for the user's Sandbox |
Sandbox usage is attributed to individual users, while Design Project usage is shared across the organization.
Inference cost of individual sandbox predictions and protein design agents vary based on the size of the biomolecules being processed.
Refer to the [Pricing Page](https://boltz.bio/pricing) for detailed cost information.
# Creating Binder Specifications
Source: https://docs.boltz.bio/user-guide/design-project-prot/Creating-Binder-Specifications
Define protein binder specifications and design motifs
## Overview
Binder specifications define the protein binder properties (modality, structure, design regions) that will be used to generate candidates.
## Creating a New Binder Specification
![Nanobody specification structure selection]()
1. From your project dashboard, click **"+ Add"** in the Binder Specifications section.
2. **Step 1: Setup** - Give your binder specification a name and choose the protein modality:
Design short peptide sequences
Create antibody heavy and light chains
Design single-domain antibodies
Other protein types
3. Click **"> Continue to Specification"**.
## Starting Specification
### Option 1: Structure-Based Specification
![Nanobody specification structure selection]()
**Select a template:**
* Choose from pre-loaded structures (e.g., vWF A1-Caplacizumab Complex, IL-6R-Vobarilizumab Complex)
* Each template shows PDB ID and description
**Or import your own:**
* Enter a **PDB ID** and **Assembly ID** (default: 1), then click **"Import"**
* Or **upload a structure file** (.cif, .cif.gz, etc.)
### Option 2: Sequence-Based Specification
Click the **"Sequence"** tab to add sequences directly instead of using a structure.
## Select Residues
![Select residues for nanobody specification]()
Select the residues you want to include in your design:
* **Recommended:** ≤300 residues
* Use **CDR1**, **CDR2**, **CDR3** quick-select buttons for antibodies/nanobodies
* Or manually select by clicking and dragging on the sequence
* Selected residues are highlighted in green
* The 3D structure updates to show selected regions
**Numbering schemes:**
* **Chothia** (default) or **Kabat** numbering for antibodies/nanobodies
* Switch between schemes using the tabs above the sequence
Click **"> Continue to Design"** when done.
## Design Configuration
Configure design motifs and constraints:
![Nanobody specification structure selection]()
### Excluded Amino Acids
Specify amino acids that will not be used in designed regions (e.g., exclude Cysteine to prevent unwanted disulfide bonds).
![Nanobody specification structure selection]()
### Create Design Motifs
1. **Select residues** on the sequence or 3D structure
2. Right-click or use the context menu to **"Replace with design motif"**
3. For CDR regions, you can specify:
* **Fixed length** - Replace with same number of residues
* **Variable length** - Specify a range (e.g., 10-25 residues for CDR3)
![Nanobody specification structure selection]()
![Nanobody specification structure selection]()
![Nanobody specification structure selection]()
### Motif Management
* Each motif is color-coded and shown in the sequence viewer
* Click a motif to edit it
* Click the **X** on a motif to remove it
* Summary shows total residues and number of motifs
Click **"> Create Specification"** when ready.
## After Binder Specification Creation
Once your binder specification is created, you'll see options to:
* **Generate binders with AI** - Start a virtual screen using this binder specification
* **Create Another Target** - Add another target to design binders against
* **Skip for now** - Continue later
# Creating Experiments
Source: https://docs.boltz.bio/user-guide/design-project-prot/Creating-Experiments
Set up hypothesis-driven virtual screening campaigns
## Overview
Experiments organize your virtual screens around specific hypotheses. Each experiment can contain multiple virtual screens and helps you track progress toward your research goals.
## Creating a New Experiment
1. From your project dashboard, click **"+ New Experiment"** in the Experiments section.
![Create new experiment form]()
2. Enter a **Name** (required) - e.g., "First campaign to find initial hits"
![Create new experiment form]()
3. Enter a **Hypothesis** (optional) - Describe the specific goal or hypothesis you're testing
Good hypotheses are specific and actionable. For example: "Identified Pocket A on target X as having potential for blocking target X activity through high-affinity protein binder engagement."
4. Click **"Create Experiment"**.
## Experiment Dashboard
Once created, you can access your experiment to:
* View all virtual screens within the experiment
* See candidate results and metrics
* Triage and organize candidates
* Add new virtual screens or quick-add candidates
## Adding Candidates to Experiments
From an experiment, you have two options:
![Create new experiment form]()
### Virtual Screen
![Create new experiment form]()
Generate or upload thousands of candidates to evaluate against your target(s). Best for large-scale screening campaigns.
### Quick Add Candidates
Add up to 20 candidates manually to dock and evaluate against your target(s). Useful for testing specific sequences or small batches.
# Creating Targets
Source: https://docs.boltz.bio/user-guide/design-project-prot/Creating-Targets
Define protein targets for binder design
## Overview
Targets are the protein structures you want to design binders against. You can define targets by entering sequences or uploading existing structure files.
## Creating a New Target
1. From your project dashboard, click **"+ Add target"** in the Targets section.
2. Enter a **Target Name** (required).
![Create new target form]()
3. Choose how to define your target:
Click **"+ Polymer"** then **"Protein from Sequence"**
Paste or enter your protein sequence (maximum 1,300 residues)
Click **"Import"**
After submitting, Boltz will predict the structure automatically
Upload a structure file (.cif, .cif.gz, etc.)
Structure prediction step is skipped
4. Click **"Continue"** to proceed to target initialization.
## Target Initialization
After creating a target, you'll go through a 4-step initialization workflow:
### Step 1: Verify Unbound Structure
![Create new target form]()
If you use sequence to define target, Boltz predicts the structure of the target. Review the predicted unbound structure, especially around the planned binding site. Only use targets with well-predicted structures.
**Optional:** Add constraints (Bond, Contact, Pocket) to induce conformational changes.
### Step 2: Select Crop
![Create new target form]()
Select the residues you want to keep. Ideal crop size is **less than 350 residues** (maximum 700 residues).
* Use the sequence viewer to select residue ranges
* The 3D structure updates in real-time to show the cropped region
* Click **"Select All"** to include all residues, or manually select ranges
### Step 3: Select Epitope (Optional)
![Create new target form]()
Select the residues on your target that you want your designs to bind to.
* This step is optional—skip if you don't want to specify an epitope
* Selected residues are highlighted in green in both sequence and 3D views
* Click **"Reset"** to clear selections
### Step 4: Select Flexible Residues (Optional)
![Create new target form]()
Select residues that the design model should treat as flexible during design.
* This step is optional—skip if the target should be treated as rigid
* Flexible residues allow for conformational changes during binding
* Selected residues are highlighted in green
Click **"Complete Setup"** to finish target initialization.
## After Target Creation
Once your target is created, you'll see options to:
* **Discover New Binders** - Create a binder specification for your virtual screen
* **Upload candidates from CSV** - Evaluate known candidates using AI
* **Skip for now** - Add the binder specification later
# Overview
Source: https://docs.boltz.bio/user-guide/design-project-prot/Overview
Collaborative workspaces for protein binder design
Design Projects are shared workspaces where team members within an organization can collaborate on protein binder design projects. They facilitate hypothesis-driven phases of a research project and provide a space to collate knowledge, designs and predictions in an organized and accessible hierarchy. By default, all work is visible to everyone within your organization.
## Project Hierarchy
Worked on by a team of scientists, optimizing protein binders against one or more **Targets**. Contains **Experiments** as sequential or parallel phases of the design cycle.
The protein structure you want to design binders against. Can be defined by sequence or uploaded structure.
Defines the protein binder specifications (e.g., nanobody, antibody, peptide) and design motifs.
A hypothesis-driven collection of **Virtual Screens** used to assess binding poses and affinity scores.
## Home
![Boltz platform home page showing organization overview and projects list]()
Your home directory is the main point of entry, where you have access to:
1. **Organization:** Your organization account where you invite members, set up billing, etc.
2. **Sandbox:** A private workspace to test hypotheses and explore protein structures before scaling up.
3. **CLI:** Submit and manage prediction jobs using our command-line interface.
4. **Projects:** All your design projects are listed here.
### Project Cards
Each project card displays:
* **Project title** and **category** (Protein/Small Molecule)
* **Last updated** timestamp
* **Metrics:** Number of targets, experiments, and candidates
Click any project card to navigate to the project dashboard.
## Project Dashboard
![Project dashboard showing targets, libraries, tags, and experiments]()
The project dashboard provides access to:
* **Targets:** List of protein targets added to the project
* **Binder Specifications:** Defines what proteins you want to design (modality, structure, design regions)
* **Tags:** Labels used to organize and categorize candidates
* **Experiments:** All experiments conducted in this project
### Getting Started Workflow
New projects include a guided workflow (typically 4 steps):
1. **Add your first target** - Create a target to start your project
2. **Add your binder specification** - Define what proteins you want to design
3. **Create your experiment** - Set up an experiment to run virtual screens
4. **Add/generate candidates** - Add candidates to your experiment
## Experiment Examples
Good experiment hypotheses are specific and actionable:
**Name:** Designing a de novo protein binder for Pocket A of Target X
**Hypothesis:** Identified Pocket A on target X as having potential for blocking target X activity through high-affinity protein binder engagement. Screening 60,000 generated molecules to find the initial hit for a drug discovery campaign.
**Name:** Antibody lead optimization targeting Epitope B on Target X
**Hypothesis:** Identified Epitope B on Target X as a functional site for pathway blockade, and hypothesized that optimizing the lead antibody's CDR interactions with this epitope would increase binding affinity and inhibitory activity.
# Quickstart: Protein Design Complete Workflow
Source: https://docs.boltz.bio/user-guide/design-project-prot/Protein-Design-Example
Step-by-step example of designing protein binders
This guide walks through creating a complete protein binder design project from scratch as a single workflow.
For further information and discussion on each step, please see the sectional breakdown within this documentation.
## Step 1: Create a New Project
![Create new project dialog]()
1. From the home page, click **"New Project"**
2. Enter a project name (e.g., "Designing for external domain of EGFR")
3. Select **"Protein"** modality
4. Click **"Create"**
![Create new project dialog]()
## Step 2: Add Your First Target
![Create new target form]()
1. From the project dashboard, click **"+ Add target"** in the Targets section
2. Enter target name (e.g., "Target-EGFR")
3. Choose **"Define Sequences"** tab
4. Click **"+ Polymer"** then **"Protein from Sequence"**
5. Paste your protein sequence (or import from UniProt)
6. Click **"Import"** then **"Continue"**
### Initialize Target
Complete the 4-step initialization:
1. **Verify Unbound Structure** - Review the predicted structure
![Verify unbound structure step]()
2. **Select Crop** - Choose region of interest (less than 350 residues ideal)
![Select crop region step]()
3. **Select Epitope** (Optional) - Mark binding site residues
![Select epitope residues step]()
4. **Select Flexible Residues** (Optional) - Mark flexible regions
![Select flexible residues step]()
Click **"Complete Setup"** when done.
## Step 3: Create a Binder Specification
![Nanobody specification structure selection]()
1. Click **"+ Add"** in the Binder Specifications section
2. Enter binder specification name (e.g., "Nanobody design V1")
3. Select **"Nanobody"** modality
4. Click **"> Continue to Specification"**
### Specify Structure
![Specify Structure]()
Choose one:
* **Pre-loaded template** - Select from provided structures e.g., vWF A1-Camplacizumab Complex
* **PDB ID** - Import via PDB ID (e.g., 7EOW)
* **Upload file** - Upload your own structure
### Select Residues
![Select residues for nanobody specification]()
* All residues are selected by default.
* You can also manually select residues by clicking/dragging
* Recommended: ≤300 residues
### Configure Design
1. **Exclude amino acids** (e.g., Cysteine)
2. **Create design motifs:**
* Select CDR regions
* Right-click then **"Replace with design motif"**
* For CDR3, specify variable length (e.g., 10-25 residues)
3. Click **"> Create Specification"**
![Design configuration with motifs]()
![Design configuration with motifs]()
![Design configuration with motifs]()
![Design configuration with motifs]()
## Step 4: Create an Experiment
1. Click **"+ New Experiment"** in the Experiments section
2. Enter name (e.g., "First campaign to find initial hits")
3. Enter hypothesis (optional but recommended)
4. Click **"Create Experiment"**
![Create new experiment form]()
![Create new experiment form]()
## Step 5: Run Virtual Screen
1. From the experiment, click **"New virtual screen"**
2. Enter screen name (e.g., "Small virtual screen to test the setup")
3. Select **"Design (Generative)"** type
4. Set budget:
* Start with **30-100** for a pilot
* Scale to **10k-100k** for production
5. Select your binder specification
6. Select your target
7. Select your experiment
8. Click **"Start Protein Design"**
![New protein design screen form]()
![New protein design screen form]()
## Step 6: Monitor Progress
![Virtual screen running status]()
While running, you can:
* View progress and elapsed time
* Pause or stop the screen
* Click **"View Results in Experiment"** to see candidates as they're generated
## Step 7: Evaluate Results
Once completed:
1. **Review metrics:**
* Top performers binding confidence graph
* Candidates above threshold counts
![Virtual screen completed results]()
2. **View candidates:**
* Switch to **"Table"** tab
* Filter and select candidates
* Review 3D structures in the viewer
* Compare properties (ipTM, pLDDT, binding confidence)
![Triage view with detailed candidate information]()
3. **Triage:**
* Use thumbs up/down and flags
* Tag promising candidates
* Export for experimental validation
Start with a small pilot (50-100 designs) to validate setup
Scale up to 10k-100k designs for production campaigns
Review top performers first, then explore diversity
Use tags to organize candidates by different criteria
Export promising candidates for experimental validation
# Results and Triage
Source: https://docs.boltz.bio/user-guide/design-project-prot/Results-and-Triage
Evaluate and organize protein binder candidates
## Overview
After a virtual screen completes, candidates are available in your experiment for evaluation, triage, and further analysis.
## Experiment View
![New protein design screen form]()
The experiment view provides three tabs:
* **Overview** - Summary statistics and metrics
* **Table** - List view of all candidates with filters and sorting
* **Triage** - Organized view for candidate evaluation
## Candidate Table
Each candidate row shows:
* **Sequence** - Protein sequence with CDR regions highlighted (for antibodies/nanobodies)
* **Id** - Unique identifier (e.g., NB-X7CMZA38)
* **Triage** - Thumbs down, flag, and thumbs up counts
* **Created By** - Which virtual screen generated this candidate
* **Created** - Timestamp when the candidate was generated
* **Binding Confidence** - Predicted binding affinity score for the target
* **Structure Confidence** - Overall structure quality score
### Filtering and Selection
* **Default view** dropdown - Pre-configured filter sets
* **Columns** button - Show/hide table columns
* **Filters** button - Add custom filters including:
* **Text filters** - Filter by Id, Created By, etc.
* **Number filters** - Filter by ipTM, pLDDT, Binding Confidence, Structure Confidence, Loop Fraction using sliders
* Filter chips show active filters and can be removed individually
* **Select candidates** - Use checkboxes to select multiple candidates
* **Download** - Export selected candidates
### Bulk Actions
When candidates are selected:
* **Add to experiment** - Move to another experiment
* **Tag** - Apply tags for organization
* **Download** - Export candidate data
## 3D Visualization
The right panel shows:
### Target Selection
* Dropdown to switch between targets
* Selected candidate IDs displayed as colored tags
* Download and fullscreen options
### 3D Structure Viewer
* Interactive 3D ribbon diagram showing:
* **Target** (light grey)
* **Binder candidates** (colored by selection)
* 3D axis indicator for orientation
* Rotate, zoom, and pan controls
### Properties Comparison
Compare metrics across selected candidates:
* **ipTM** - Interface predicted Template Modeling score
* **pLDDT** - Predicted Local Distance Difference Test (structure confidence)
* **Binding Confidence** - Predicted binding affinity
* **Structure Confidence** - Overall structure quality
Each chart shows all selected candidates as colored lines, making it easy to identify top performers.
## Triage View
The **Triage** tab provides a detailed view for evaluating individual candidates:
### Left Panel - Candidate Details
* **Sequence Viewer:**
* Full protein sequence with Framework Regions (FR) and Complementarity Determining Regions (CDR) highlighted
* CDR regions color-coded (CDR1, CDR2, CDR3)
* Residue numbers displayed above sequence
* Navigation arrows to move between candidates
* **Candidate Actions:**
* **Star** - Favorite a candidate
* **Thumbs down** - Reject candidate
* **Flag** - Mark for further review
* **Thumbs up** - Approve candidate
* **+ Tag** - Apply tags for organization
* **Properties and Descriptors:**
* **Properties tab** - View metrics (ipTM, pLDDT, Binding Confidence, Structure Confidence)
* **Descriptors tab** - View structural features (Loop Fraction, Helix Fraction, Sheet Fraction)
* Switch between **Table** and **Plot** views for each
### Right Panel - 3D Structure and Comparison
* **3D Structure Viewer:**
* Shows the selected candidate's binding pose with the target
* Binder colored orange, target colored light grey/green
* Interactive controls for rotation, zoom, and pan
* **Properties Comparison:**
* Parallel coordinates plot showing all candidates
* Selected candidate highlighted in purple
* Non-selected candidates shown in dark grey
* Compare across multiple properties simultaneously
## Triage Workflow
1. **Select candidates** - Use checkboxes or filters to narrow down
2. **Review 3D structures** - Examine binding poses in the viewer
3. **Compare properties** - Use charts to identify high-scoring candidates
4. **Triage actions:**
* **Thumbs down** - Reject candidate
* **Flag** - Mark for further review
* **Thumbs up** - Approve candidate
5. **Tag candidates** - Apply tags for organization (e.g., "high-affinity", "interesting-pose")
6. **Export** - Download selected candidates for further analysis
## Tips
* Start by filtering to top performers (high binding confidence)
* Review 3D structures to ensure poses are reasonable
* Compare multiple candidates side-by-side using the properties charts
* Use tags to organize candidates by different criteria
* Export promising candidates for experimental validation
# Tips
Source: https://docs.boltz.bio/user-guide/design-project-prot/Tips
Best practices for protein binder design
Start with a small pilot (30-100 designs) to validate your setup, specifications, and predictions. Once everything looks correct, scale up to 10k-100k designs per campaign—the larger the better, especially for difficult targets or modalities.
Keep target crops under 350 residues for optimal performance. Select epitope residues to guide binding, and mark flexible regions if you expect conformational changes.
For antibodies/nanobodies, focus design on CDR regions. Use variable-length CDR3 (10-25 residues) to increase diversity. Exclude problematic amino acids (e.g., Cysteine) to prevent unwanted interactions.
Review top performers first by binding confidence, then examine 3D structures to ensure poses are reasonable. Use tags to organize candidates by different criteria (affinity, pose quality, diversity).
# Virtual Screening
Source: https://docs.boltz.bio/user-guide/design-project-prot/Virtual-Screening
Run AI-powered protein design campaigns
## Overview
Virtual screens generate protein binder candidates using AI. You can pause and resume screens at any time.
## Creating a New Virtual Screen
![New protein design screen form]()
1. From an experiment, click **"New virtual screen"** or **"Virtual Screen"**.
2. **Name** - Enter a name for this protein design screen (e.g., "Small virtual screen to test the setup")
3. **Type** - Choose your screen type:
Use AI to generate protein binders for your target. This is the most common option for discovering new candidates.
Upload CSV with protein sequences, SMILES, or other molecular data to evaluate existing candidates.
4. **Budget** - Set the maximum number of protein designs to generate:
Budget below 5k may not be sufficient to find good candidates. Start with a small pilot (50-100 designs) to validate your setup, then scale up to 10k-100k designs for production runs.
* Use the slider to select: 10, 1k, 5k, 25k, 125k, or 625k
* Estimated cost is displayed below
5. **Design Specification** - Select a binder specification to use for generating protein designs
* Shows binder specification details: Chains, Motifs, Designed residues
* Only binder specifications added to your project are shown
6. **Target** - Select the target to design binders against
* Shows target details: number of chains and residues
* Only initialized targets are available
7. **Experiment** - Select an experiment for this virtual screen
* Choose from existing experiments or create a new one
* If no experiments exist, you'll be prompted to create one
8. Click **"Start Protein Design"** to begin.
## Monitoring Virtual Screens
![New protein design screen form]()
While a screen is running, you can:
* **View progress** - See "Waiting for first candidates to be generated..." with elapsed time
* **Pause** - Temporarily pause the screen
* **Stop Run** - Cancel the screen entirely
* **View Results in Experiment** - Navigate to see candidates as they're generated
### Status Indicators
* **RUNNING** (blue) - Screen is actively generating candidates
* **COMPLETED** (green) - Screen has finished generating all candidates
* **PAUSED** - Screen is paused and can be resumed
## Viewing Results
![Virtual screen completed results]()
Once completed, the screen overview shows:
### Performance Metrics
**Top Performers: Binding Confidence**
* Tracks the best, 10th best, and 100th best candidates over time
* Shows how binding confidence improves as more candidates are processed
**Candidates Above Binding Confidence Thresholds**
* Cumulative count of candidates exceeding threshold values (>0.5, >0.6, >0.7, >0.8)
* Helps assess overall screen quality
### Screen Details
* **Status:** COMPLETED
* **Total Generated:** Number of candidates created
* **Added to experiment:** Number automatically added (top performers)
* **Settings:** Binder specification modality, chains, candidate budget
Click **"View Results in Experiment"** to see all candidates and begin triage.
# Design Tab
Source: https://docs.boltz.bio/user-guide/design-project-sm/Design-Hub
The Design Tab allows de-novo edits of molecules within the UI.
Molecules in the Design tab are cached and persist for the current Experiment/Virtual Screen, so you can build up a collection of designs across multiple sessions.
The Design tab currently supports manual molecule input using the built-in Ketcher drawing tool. Additional capabilities will be deployed in future releases.
### Adding Molecules to Design Tab
There are three ways to add molecules from your virtual screen results to the Design workspace:
![Design1]()
**Method 1: Hover over molecule**
* In the **Table** view, hover over any molecule structure
* A pen icon (✏️) appears with "Design from this template" tooltip
* Click to add the molecule to the Design tab
**Method 2: Select from table**
* Use the checkbox to select one or more molecules in the Table
* Click **Design from this template** in the top menu
* Choose to either:
* Add to existing designs already in the Design tab
* Replace existing molecules in the Design tab
**Method 3: Switch and go!**
* Switch to the Design tab and start drawing - it's that simple!
***
### Design Tab Interface
![Image]()
The Design workspace is divided into several sections:
Full-featured molecular drawing interface for creating and editing structures:
* Draw new molecules from scratch
* Edit molecules added from results
* Create copies and variants
* Standard drawing tools (bonds, atoms, rings, functional groups)
* Multiple molecules can be worked on simultaneously
Displays the Boltz prediction and pocket for selected molecules:
* Target selection dropdown (if multiple targets exist)
* Interactive 3D structure visualization - highlight sidechains and residues
* Rotatable and zoomable view
Two radar charts showing molecular properties for all molecules in the workspace. **These update live as you edit structures in the Ketcher editor.**
**Standard descriptors:**
* Exact MW (Molecular Weight)
* CLogP (Lipophilicity)
* TPSA (Topological Polar Surface Area)
* Lipinski HBD (Hydrogen Bond Donors)
* Lipinski HBA (Hydrogen Bond Acceptors)
* FSP3 (Fraction sp3 carbons)
**Additional descriptors:**
* Rotatable Bonds
* HAC (Heavy Atom Count)
* Heteroatom count
* Stereocenters
* Unspecified Stereo
* Ar Rings (Aromatic Rings)
Each molecule is color-coded for easy comparison.
Shows all molecules queued for submission:
* Thumbnail structures with IDs
* Multiple molecules can be submitted together
* Visual confirmation before submission
***
### Working with Molecules
Either add molecules from your results (using "Design from this template") or draw new molecules using the Ketcher editor.
Use the Ketcher drawing tools to modify structures - add substituents, change ring systems, modify functional groups.
The radar plots update live as you edit - watch how your modifications affect molecular properties in real-time. Each molecule's profile is overlaid for direct comparison.
Verify all molecules appear correctly in the Molecule Preview panel at the bottom right.
Click **Submit Molecule** to proceed to target selection and add these designs to your virtual screen.
***
### Submitting Designs
![Design3]()
When you click **Submit Molecule**, you'll be prompted to choose which target(s) to predict structures and binding against. You can select multiple targets simultaneously if you want to evaluate your designs against several targets.
To **track progress** of designed molecules, a bar appears under each molecule in the Molecule Preview panel, showing prediction status in real-time
Once predictions complete, your manually designed molecules are added to the virtual screen results.
***
### Finding Your Designs in Results
After submission, locate your manually created designs in the results:
**Filter by creator:**
* In the Table view, use the **Created By** column
* Filter or sort to show molecules you submitted via the Design tab
**Filter by date:**
* Use date filters to show recently submitted designs
* Helpful for distinguishing manual designs from generative campaign outputs
The Tag function can also be used to mark your manually designed molecules immediately after submission to easily distinguish them from other results.
***
### Use Cases
Select a hit from your results, then use the Design tab to manually create analogs by modifying specific substituents or functional groups based on your medicinal chemistry expertise.
Draw a small set of closely related analogs (e.g., systematic R-group exploration) and submit them together for rapid SAR assessment.
Use the radar plots to guide modifications - adjust structures to shift CLogP, reduce MW, or improve other properties while monitoring changes in real-time.
Sketch specific molecules to test structural hypotheses that may not emerge from generative campaigns. Submit against multiple targets simultaneously to assess selectivity or cross-reactivity.
***
### Best Practices
Work with 2-3 molecules simultaneously to compare property profiles directly on the radar plots
Apply tags to manually designed molecules for easy tracking in the results table
Take advantage of live property updates - modify structures and immediately see how CLogP, MW, and other descriptors change
Review binding predictions of submitted designs, then create refined versions in new Design sessions
***
# Purchasing Small Molecules
Source: https://docs.boltz.bio/user-guide/design-project-sm/Purchasing-Small-Molecules
How to order synthesizable compounds from your virtual screening results via Enamine
***
## Overview
After completing your virtual screening campaign and identifying promising candidates, you can seamlessly transition from in silico predictions to compound procurement.
The small molecule agent within Boltz Lab searches Enamine's synthesis-on-demand services, any molecule generated can be purchased from the **Enamine REAL Space** (\~75B compounds).
All generative screens using "Enamine REAL Space" chemical space filtering produce molecules that are theoretically synthesizable through Enamine's building block library.
If you have you have your own synthesis team, you can synthesize molecules yourself. Using the small molecule agent does not obligate purchasing.
***
## Workflow: From Results to Purchase
Use the Table or Triage UI to collaboratively evaluate your virtual screen output:
* **Upvote** molecules you favor based on predicted binding and properties
* **Flag** molecules for team discussion or further analysis
* **Tag** molecules with contextual labels (e.g., "high-affinity", "lead-like")
Use keyboard shortcuts for rapid triage: **←→** to navigate, **3** for upvote+flag, **2** for flag, **1** for downvote
After triage and team review, mark your final selection as **'active'** by clicking the ⭐ star icon:
* In the **Table view**: Click the star in the "ID and Tags" column
* In the **Triage UI**: Use the star action button
Active molecules are automatically added to your Design Project **Catalog**.
Navigate to the **Catalog** from your Design Project dashboard to view all active molecules across your experiments.
![Design project catalog showing selected molecules]()
The \*\*Catalog \*\*functions as your curated shortlist of molecules for procurement and provides the same filtering and analysis tools as the Results Table.
From the Catalog:
1. **Select molecules** using the checkboxes (or select all)
2. Click the **Download** button in the top menu
3. Export as **TSV** format
The file will contain SMILES strings and all associated prediction data for your selected molecules.
Send your exported SMILES strings to Enamine using the contact information below. Include:
* **SMILES strings** from your downloaded file
* **Quantities** required for each compound
* **Purity specifications** (if applicable)
* **Timeline** for delivery
Enamine will provide a quote for synthesis and timeline estimates.
***
## Enamine Contact Information
**Enamine US Inc.**
Phone: (732) 274 9150
Toll Free: +1 888 745 6132
Email: [sales\_usa@enamine.net](mailto:sales_usa@enamine.net)
**Enamine LV SIA**
Phone: +371 29 473 022
Email: [sales\_europe@enamine.net](mailto:sales_europe@enamine.net)
**Enamine LTD**
Phone: +380 44 333 5601
Email: [info@enamine.net](mailto:info@enamine.net)
***
## Important Notes
**Enamine REAL Space vs. Website:**
Boltz Lab screens the complete **\~75 billion compound** Enamine REAL virtual space. The Enamine website displays only a subset of this library.
**If you cannot find your molecule on the Enamine website, this is normal.** Email Enamine directly with your SMILES strings to confirm availability and obtain accurate quotes.
**Synthesis Timeline:**
Most compounds from the REAL Space can be synthesized on-demand within 3-6 weeks, depending on complexity and quantity. Enamine will provide specific timelines when you request a quote.
**Ordering Best Practices:**
* Start with small quantities (2-10 mg) for initial validation experiments
* Request standard purity (≥90%) unless high purity (≥95%) is essential for your assays
* Order in batches to optimize shipping costs
* Plan ahead - allow 4-8 weeks from quote to delivery for typical orders
***
## Alternative Procurement Options
### Custom Library Screening
If you have preferred compound vendors or internal libraries:
1. Export molecules from your Catalog as CSV
2. Share SMILES strings with your preferred synthesis provider
3. Many CRO partners can synthesize compounds from SMILES specifications
### Cross-Vendor Validation
For high-priority leads:
1. Request quotes from multiple synthesis providers
2. Compare pricing, timelines, and purity specifications
3. Consider ordering duplicates for critical validation studies
***
## Troubleshooting
While most REAL Space compounds are synthesizable, some edge cases may require:
* Modified building blocks (may increase cost/timeline)
* Alternative synthesis routes (discuss with Enamine)
* Selecting a structurally similar analog from your results
Enamine will advise on feasibility when you request a quote.
Yes! If you uploaded a custom library or screened pre-computed vendor libraries:
* Check the original source of your molecules
* Contact the vendor directly with SMILES or compound IDs
* Many vendors offer direct purchasing through their websites
We recommend maintaining a spreadsheet with:
* Boltz molecule ID
* SMILES string
* Enamine quote reference
* Order date and expected delivery
* Batch/lot numbers upon receipt
You can also use Tags in the Catalog to mark "Ordered", "In Transit", "Received" for tracking within Boltz Lab.
***
## Next Steps
After receiving your compounds:
1. **Experimental Validation**: Run biochemical or cell-based binding assays
2. **Hit Expansion**: Use validated hits as templates for generative design or manual optimization in the Design Hub, or via CSV uplaod
3. **Iterate**: Run follow-up virtual screens focused on validated chemotypes
Have questions about compound procurement? Contact our support team at [support@boltz.bio](mailto:support@boltz.bio)
# Results & Triage
Source: https://docs.boltz.bio/user-guide/design-project-sm/Results-and-Triage
Analyze virtual screen results and collaboratively prioritize molecules
## Results Table
Navigating to an **Experiment** shows the results table from all Virtual Screens it contains. The Table UI is divided into three sections: a data table, MolStar viewer, and a progressive flow chart of affinity predictions and RDKit descriptors.
![Image]()
### Table Features
| Feature | Description |
| --------------------- | -------------------------------------------------------------------------------------------------------------------------- |
| **Selecting a row** | Loads molecule into the viewer. Clicking the checkbox, multiple (up to 10) molecules can be selected and overlaid. |
| **ID and Tags** | Unique ID for the molecule. Click the ⭐ to mark as 'active' in the Catalog. Tags allow labelling with contextual keywords. |
| **Alert Level** | Indicates if the molecule hits any structural alert filters. |
| **Triage Elements** | Upvote/Downvote + Flag buttons for annotation. Connected to the Triage UI. |
| **Boltz Metrics** | Metrics for each prediction, grouped by target if multiple exist. |
| **RDKit Descriptors** | Standard molecular descriptors for triage and analysis. |
| **Columns + Sorting** | Toggle columns, drag to rearrange, click headers to sort. |
| **Filters** | Numeric sliders and text/contextual filters available. |
| **Download** | Export all molecules and metrics from the table. |
Results are downsampled to a couple thousand at most for user convienience.\
\
Sorting by a column will display to top results, with additional filters then applied to show the highest fidelity hits
Download will provide an output data file of all molecules in the downsample, or remaining after filters are applied.\
\
Selecting molecules will only download the data for those molecules.\
\
Please contact support for any additional download requests
***
### Boltz Scoring Metrics
Primary scoring function for **hit discovery**. Continuous score between 0 and 1, with 1 being highest confidence of binding.
This is similar to the affinity probability output of Boltz-2.
Scoring function for **lead optimization** settings where relative ranking within a series matters. Continuous score between 0 and 1, with 1 being the strongest binder.
This is similar to the affinity value output of Boltz-2.
Measure of confidence on the correctness of the structure prediction, combines pLDDT and ipTM score.
Continuous score between 0 and 1, with 1 being highest confidence.
### Viewer Features
Note that, for large targets, the model focuses the prediction in the pocket selected in the Target creation stage. If you are looking for the molecules to bind to a different pocket, add a new target. If you are looking to analyze the specific conformational changes on the full target from a ligand binding, use the Sandbox.
| Feature | Description |
| ---------------------- | ------------------------------------------------------------------------- |
| **Multiple selection** | Up to 10 molecules can be overlaid. Selecting more displays the first 10. |
| **Click molecule** | Loads nearby pocket residues and highlights proposed interactions. |
| **Download** | Top-right button exports 3D structures of loaded molecules. |
### Progressive Flow Chart
This 2D plot visualizes relative scores across various Boltz affinity metrics or RDKit descriptors within the screen.
![Progressive flow chart filtering]()
Click and drag over a vertical column to apply filter ranges to the dataset. Highlighted lines show molecule trajectories through these filters. The table updates to show only molecules within the filtered ranges.
Hover over a line to display the molecular structure and ID in the tooltip.
***
## Triage UI
The **Triage** UI allows collaborative or independent assessment of virtual screen output with a focused, structure-centric view for each molecule design.
![Image]()
### Triage Actions
Mark molecules you favor based on available data
Mark molecules you disfavor
Flexible marking for problematic predictions, chemical motifs, or molecules warranting discussion
Mark as 'active' to add to the Catalog
### Keyboard Shortcuts
| Action | Shortcut |
| ------------------------ | ------------------ |
| Previous / Next Molecule | **← →** arrow keys |
| Upvote / Downvote + Flag | **3** |
| Flag | **2** |
| Downvote | **1** |
### Tags
Tags allow free-form labelling shared across the Design Project. Multiple tags can be added to a molecule. When creating a new tag, a longer description can be provided while keeping the tag name short for a clean interface.
Select **Tags** from the left-hand menu to review all tags later.
***
## Catalog
The **Catalog** captures all 'active' molecules within a Design Project, forming a database across all Experiments and Virtual Screens.
![Design project catalog]()
Active molecules in the Catalog are designs of particular interest to the Design Project team.
The Catalog shares the same features as the Table UI. Unchecking the ⭐ sets a molecule back to inactive and removes it from the Catalog.
### Recommended Workflow
1. Collaboratively (or independently) upvote, downvote, or flag molecules in the Table or Triage UI from a Virtual Screen
2. After all users have made triage recommendations, mark selected molecules as 'active' to add them to the Catalog
3. Use the Catalog as a shortlist for design review meetings or further analysis
# Overview
Source: https://docs.boltz.bio/user-guide/design-project-sm/Small-Molecule-Overview
Collaborative workspaces for small-molecule design within Boltz Lab
Design Projects are shared workspaces where team members within an organization can collaborate on projects.
They facilitate hypothesis-driven phases of a research project and provide a space to collate knowledge, designs and predictions in an organized and accessible hierarchy.
By default, all work is visible to everyone within your organization.
## Project Hierarchy
Boltz Lab provides a hierarchy organisation to organise Design Projects and provide structure:
Worked on by a team of scientists, optimizing small-molecules against one or more **Targets**. Contains **Experiments** as sequential or parallel phases of the DMTA cycle.
A single or collection of **Virtual Screens** used to assess Boltz binding poses and affinity scores for a particular hypothesis.
A collection of small-molecule designs for assessment against a **Target**. Can range from small bespoke designs to large commercial libraries.
A Design Project will contain multiple experiments, and an experiments will contain one or more Virtual Screens to fulfill its aim.
***
## Creating a Design Project
Click **Design Projects** to get started.
![Image 2]()
Provide a project name, and select whether it is for designing small-molecule or protein binders. For this section, we will focus on small molecules.
![Image]()
The in-built tutorial will guide you on how to **Get started with your project**, with additional guidance continuing in the Boltz Lab documentation
![Image]()
***
## Design Project Guidance and Tips
### Experiment Examples
Good experiment hypotheses are specific and actionable:
**Name:** Pocket A Hit Finding Campaign
**Hypothesis:** Identified Pocket A on target Kinase X as having potential for pharmacological function. Screening 60,000 generated molecules to find hit starting matter for a drug discovery campaign.
**Name:** WT Kinase Series 1
**Hypothesis:** Exploring H-bond donors at R2 position of series core to improve kinase binding affinity to wild-type Kinase A, while maintaining solubility. Enumerated library from Supplier building-block library.
**Name:** Screening of Vendor Library A
**Hypothesis:** Commercial vendors supply libraries of privileged scaffold binders for the class of proteins my Target belongs to. This screen will upload that vendor csv file, and screen 1200 high fidelity molecules for our target.
**Name:** Binding Pose Evaluation of Project Y Chemical Matter
**Hypothesis:** Adding pre-existing Project Y molecules to the platform to integrate Boltz into pipeline. Screen of back-catalog of chemical matter to fully evaluate binding poses and affinity scores using the Boltz models.
### Virtual Screen Sources
Virtual screens can come from diverse sources:
* Small, uploaded bespoke designs
* Large commercial vendor or uploaded libraries
* Generative design campaign, using Boltz AI models to propose new candidates
Once a virtual screen is complete, it can be **Triaged** within the platform.
***
# Targets
Source: https://docs.boltz.bio/user-guide/design-project-sm/Targets
Define protein targets for your design campaigns
To begin designing binders, add a target protein/receptor via the **+ Add Target** function in the **Targets** section.
***
## Adding a Target
![Image]()
Add the target name and the amino acid sequence via one of the following options:
| Import Method | Description |
| ------------------ | ------------------------------------------------------------------------------------- |
| **Sequence** | Paste the FASTA sequence and select any post-translational mutations or modifications |
| **RCSB Import** | Add sequence via the RCSB-PDB ID code of a deposited crystal structure |
| **UniProt Import** | Add sequence via the UniProt code |
| **File Import** | Add sequence from a .cif structure file |
**Targets** are not restricted to the main receptor of interest in a project. Additional off-target receptors, mutants, homologs, etc., can be added within a Design Project for Boltz structure and binding affinity assessment.
At this point, targets can only be formed by protein chains. Co-factors and RNA/DNA targets are not yet supported for small-molecule projects.
When ready, click **Create Target**. The apo Boltz predicted structure will then be generated.
![Image]()
## Verifying the Apo Structure
Verify that the unbound structure looks roughly correct. If there are particular issues, you can try solving them by adding constraints. Once satisfied with the apo pose, click **Continue**.
***
## Defining the Binding Pocket
To map out a binding pocket, either:
1. Select residues in the left-hand sequence panel, or
2. Select them directly in the 3D viewer
Selected residues are highlighted in green. If you don't know the binding pocket, you can leave it blank.
![Image]()
To help with pocket detection, Boltz uses molecular probes. You can enter a SMILES string of known binding molecules to assist this process.
Once you click **Predict pocket structure**, the model will predict the position of the pocket based on the information you provided. This process takes a couple of minutes.
![Image]()
This verification step is important because from this point onwards the model will use the identified pocket as the specific target for all molecules. If the probes are bound to the wrong pocket, return to the previous screen and select more pocket residues.
Once complete, you are ready to set up an **Experiment** and begin a **Virtual Screen**.
***
## Deleting Targets
To avoid conflicts, currently Targets that have completed setup cannot be deleted.
Targets in progress can be deleted be selecting from the three dot menu in the Target list.
![Image]()
***
# Virtual Screening - Setup
Source: https://docs.boltz.bio/user-guide/design-project-sm/Virtual-Screening
Run generative designs or screen custom libraries
Once a Target is created, Boltz Lab witll guide you through starting your first **Virtual Screen**
![Image]()
## Creating an Experiment
Follow the instruction and select **+** **Create New Experiment** fromprompts
![Image]()
Give the **Experiment** a name and a hypothesis that describes the work being carried out. The Experiment space is collaborative and accessible to other users within the organization.
**Name:** Experiment 1
**Hypothesis:** Molecules for screening
**Name:** WT Kinase Series 1
**Hypothesis:** Exploring H-bond donors at R2 position of series core to improve kinase binding affinity to wild-type Kinase A, while maintaining solubility. Enumerated library from Supplier building-block library.
To continue, click **New Virtual Screen**.
***
## Generative Virtual Screen
A **Generative** screen uses generative AI models and active learning to search molecular space and find binders for the selected Target, optimizing the predicted binding confidence.
![Download (1)]()
### Parameters
If multiple Targets are added to the Design Project, select the target you want to optimize affinity for.
By default, a generative screen searches within the expanded \~75B Enamine REAL Space. This ensures all output molecules are theoretically synthesizable. Alternatively, choose not to apply any chemical-space filter.
**Normal Filtering** (default): Curated filtering excluding most problematic molecules but allowing some motifs (phenols, anilines, aryl halides, esters) that can be optimized away in later design cycles.
**Extra Filtering**: Extends to include functional groups with chemical stability issues.
**Aggressive Filtering**: Applies aggressive filtering to remove edge cases.
Specify filters based on RDKit molecular descriptors to keep small-molecule generation within your desired molecular/physicochemical space. **Preset** filters are available for convenience.
Set filters at generation time rather than filtering afterwards to optimize compute usage.
Sets the maximum number of molecules for generation and scoring. You can pause/stop the screen at any time.
It is recommended **not** to run fewer than 20,000 molecules, as this typically does not allow convergence in the active learning model.
***
## Library Virtual Screen
**Custom** or **Pre-computed Libraries** can alternatively be screened by selecting **Library (CSV)**.
### Upload CSV
The file should contain a **SMILES** column with the molecules identifier.
![CSV upload interface]()
### Pre-computed Libraries
Pre-computed vendor compound libraries can be selected without needing to upload CSV files.
![Pre-computed library selection]()
**Cross-screening tip:** For screening molecules against multiple targets, export the desired molecule designs from your Experiment or Virtual Screen, and use them as a custom library in the same or new Experiment for cross-target/off-target Boltz predictions.
# Experiments
Source: https://docs.boltz.bio/user-guide/design-project-sm/Virtual-Screening-Monitoring
Track progress, interpret quality metrics, and manage running screens.
Once your virtual screen starts running, the platform continuously monitors prediction quality and tracks progress toward your screening objectives.
You can view live updates, respond to quality alerts, and adjust settings as needed.
***
## Experiment Overview
Selecting a **Project**, will display all \*\*Experiments \*\*within it.
![Image]()
The \*\*Overview\*\* tab provides a summary, and details all **Virtual Screens** nested within the **Experiment**.
![Image]()
## Virtual Screen Overview
Select a virtual screen to access the real-time monitoring dashboard.
![Image]()
### Status Overview
The right panel displays key information about your screen:
| Field | Description |
| -------------- | -------------------------------------------------------------- |
| **Status** | Current state: Running, Paused (awaiting review), or Completed |
| **Experiment** | Parent experiment name |
| **Created** | Start timestamp |
| **Duration** | Total runtime so far |
| **Expires in** | Days remaining before paused run is stopped (30 days) |
### Results Summary
Track screening progress in real-time:
| Metric | Meaning |
| ----------------------- | --------------------------------------------------------------------- |
| **Total Generated** | Total molecules that have been generated and scored by Boltz's agents |
| **Added to Experiment** | Molecules meeting criteria above binding confidence thresholds. |
### Settings Reminder
View your original screen configuration:
* **Objective:** The target you're optimizing binding affinity for
* **Budget:** Maximum number of molecules to generate (your computational budget)
***
## Interpreting Progress Charts
### Top Performers Chart
Tracks the binding confidence of the **best**, **10th best**, and **100th best** molecules over the course of the screen.
**What to look for:**
* **Rising green line (Best):** The best molecule keeps improving - the screen is finding better binders
* **Plateau after initial climb:** Normal behavior once the model converges on high-affinity scaffolds
* **All lines close together:** Limited chemical diversity or convergence
* **10th/100th best rising:** Good sign - you're finding multiple high-quality hits, not just one outlier
### Candidates Above Threshold Chart
Shows the cumulative count of molecules exceeding different binding confidence thresholds (**>0.5, >0.6, >0.7, >0.8**) as the screen progresses.
**What to look for:**
* **Black line (>0.5) rising steeply:** Many molecules predicted as likely binders
* **Green line (>0.8) accumulating:** High-confidence hits being discovered consistently
* **Flat lines:** Few molecules meeting thresholds - may indicate target difficulty or inappropriate chemistry
* **Only >0.5 line rising:** Moderate-confidence predictions but few strong hits
For hit discovery, prioritize **>0.7** candidates, however **>0.5** may be acceptable starting points.
***
## Managing Your Screen
Click the **three-dot menu** next to the virtual screen title to access management options:
### Available Actions
Navigate to the results and triage interface to analyze all scored molecules, regardless of whether the screen has completed. Useful for early decision-making during long screens.
Update the virtual screen name to better reflect its purpose or findings (e.g., "H-bond donor exploration" → "Series 1 SAR expansion").
Modify the computational budget (increase or decrease molecule count) and restart the screen.
**Use cases:**
* Increase budget if early results are promising and you want deeper exploration
* Decrease budget if you've already found sufficient hits and want to conserve resources
* Restart with adjusted molecular filters after reviewing initial outputs
Editing the budget and restarting will not discard the current progress.
***
## Paused vs. Running vs. Completed States
Your virtual screen will transition through different states:
| State | What It Means | What You Can Do |
| ------------- | ------------------------------------------------------------------------------------------------ | --------------------------------------------------- |
| **Running** | Actively generating and scoring molecules | Monitor progress, view live results |
| **Paused** | Automatically paused due to low ipTM alert
Manually paused to review current progress | Review & Approve to continue, or Stop Run to cancel |
| **Completed** | Reached the budget limit or manual stop | View final results, triage molecules, export data |
Even in the **Running** state, you can click "View Results in Experiment" to start analyzing molecules that have already been scored - you don't have to wait for completion.
***
## Best Practices for Monitoring
For generative screens running 25,000+ molecules, check the dashboard once per day to catch issues early.
Review the apo structure, pocket definition, and early molecule structures to diagnose the issue if chemical hits do not look as expected.
If the top performers chart plateaus and the >0.8 threshold stops accumulating new molecules, the screen has likely converged. You may not need to run the full budget.
For well-validated targets (kinases, GPCRs), expect >0.7 binding confidence scores.
After 10,000-15,000 molecules in a generative screen, emerging patterns should be representative of overall trajectory.
***
## Troubleshooting Common Issues
**Likely cause:** Poor target structure quality or incorrect pocket definition.
**Solution:** Stop the screen, return to target setup, and:
* Add constraints to fix the apo structure conformation
* Redefine the binding pocket with more residues or probe molecules
* Consider using a crystal structure template instead of predicted structure
**Likely cause:** Chemical space mismatch, overly restrictive molecular filters, or target difficulty.
**Solution:**
* Review molecular filters - remove unnecessary restrictions (MW, LogP, HBD/HBA)
* Check if chemical space filter is excluding productive scaffolds
* For challenging targets, accept lower affinity thresholds (>0.5 instead of >0.7)
**Likely cause:** Generative model has converged on optimal scaffolds within the search space.
**Solution:** This is normal - you've found the candidate binders the agent is most confident on.
Stop the screen, triage results, and consider:
* Adjusting molecular filters to explore different chemical space
* Switching to library screening with novel scaffolds
* Moving successful hits into lead optimization cycles
***
## Understanding ipTM Alerts
During screening, Boltz monitors the **ipTM (interface predicted TM-score)** for each protein-ligand prediction. ipTM measures the model's confidence in the quality of the predicted protein-ligand interface - essentially how well the model believes the binding pose is resolved.
**ipTM in brief:** A score from 0-1 indicating the model's confidence that the protein-ligand interface structure is accurate, with higher scores meaning more reliable predictions.
### Low ipTM Score Alert
If the average ipTM score drops below the quality threshold (typically **0.7**), the platform automatically pauses your virtual screen and sends you an alert.
![Iptm]( "Iptm")
**Why you received this alert:**
Low ipTM scores suggest the model is struggling to generate confident protein-ligand binding predictions. This could be caused by:
The model's understanding of structure may be poor, preventing proper ligand docking
Cryptic pockets, allosteric sites, or unusual binding modes that are difficult for the model to understand
Early molecules in generative screening may be exploring unsuitable chemical space
### What Should You Do?
You have two options when you receive a low ipTM alert:
**When to continue:**
* Target is inherently challenging (flexible binding sites, allosteric pockets) and low ipTM is expected
* You have high confidence in your apo structure and pocket definition
* You're willing to accept some lower-quality predictions to gather more data
The screen will resume from where it paused.
**When to stop:**
* You suspect the target structure or pocket definition is incorrect
* The candidate molecules are not appropriate for your target (wrong chemistry)
* You want to revisit constraints or pocket specification before continuing
Stop the run, review your target setup, and restart with adjustments.
# Quickstart
Source: https://docs.boltz.bio/user-guide/introduction/Quickstart
Signed up to Boltz Lab and want to begin predicting? This is the place for you.
This documentation gives an overview of all the features and capabilities that Boltz Lab has to offer. \
\
**TLDR?** - We completely understand!\
\
We have tried to design the platform to be easy to use. So getting started with your first prediction is a good way of learning as you go.
This Quickstart guide will get your first prediction submitted as quickly as possible.
***
## Essential: Sandbox vs Design Projects
![Image]()
Not sure which to use? Here's a quick comparison:
| Feature | Sandbox | Design Projects |
| :---------------------- | :---------------------------- | :--------------------------------------- |
| **Best for** | Quick experiments, validation | Systematic campaigns, team collaboration |
| **Visibility** | Private (you only) | Shared with team |
| **Virtual Screening** | ✗ | ✓ (thousands of molecules) |
| **Generative Design** | ✗ | ✓ (AI-powered molecular generation) |
| **Affinity Prediction** | ✓ | ✓ |
| **Constraints** | ✓ | ✓ |
| **Results Storage** | Personal history | Project-organized |
Start in Sandbox to validate your target and binding site, then move to a Design Project when you're ready to screen compound libraries, run generative campaigns, or simply manually test different hypotheses.
***
## Quick Start
Navigate to **Sandbox** from the Welcome box, or from the top menu bar
Click **+ New Prediction** to launch the prediction wizard
Import your protein (from PDB or sequence) and ligand (SMILES, structure, or CCD code)
Add bond, contact, or pocket constraints to guide the prediction
Click **Submit Prediction** and results typically arrive in 1-3 minutes
# Register / Log In to Boltz Lab
Source: https://docs.boltz.bio/user-guide/introduction/Register--Log-In-to-Boltz
## Sign-up
Boltz Lab is currently in beta release, you can request access here: XXX.
First-time sign-up and log-in instructions will be provided by email.
If you don't already have access, please select that option, and complete the form. A member of the Boltz team will be in contact with further instructions.
![Image]()
Once your account has been created, please create or join an organization for your organization or team.
![Image]()
If you are unsure, please consult your organization's super-user / administrator, or Boltz for further details.
# Quick-Start Video
Source: https://docs.boltz.bio/user-guide/introduction/Tutorial
A quick 101 to get you started!
We have prepared a quick overview for getting started with the sandbox projects utilizing the Boltz models and agents within Boltz Lab.\
\
More tutorial videos are coming soon!
# Welcome to Boltz Lab
Source: https://docs.boltz.bio/user-guide/introduction/Welcome-to-Boltz
VideoBuilt on top of Boltz's state-of-the-art AI agents, the Boltz Lab enables structure-based small-molecule and protein design to any target.
The agents are powered by the Boltz model ecosystem, frontier AI models to predict complex structures, binding affinities and designing new binders.
Get started with Boltz Lab
Collaborative campaigns for hit-discovery and lead optimization
Collaborative campaigns for de-novo protein design and optimization
Need help? Contact our [support team](mailto:support@boltz.bio) or join the [Slack community](https://boltz.bio/join-slack).
# Example Bond: Covalent Inhibitors
Source: https://docs.boltz.bio/user-guide/sandbox/Constraints/Example-Bond-with-Covalent-Inhibitors
Guide to modeling covalent inhibitors using the Bond constraint in Sandbox mode
***
## Introduction
This guide will use the covalent inhibitor AMG 510 bound to KRAS G12C (PDB Code 6OIM) as an example prediction.
Modeling of covalent inhibitors is enabled within Boltz Lab, and currently limited to the Sandbox mode.
To specify a covalent bond, use the \*\*Bond \*\*constraint option on the \*\*Constraints \*\*page. This allows you to define which atoms should form a covalent connection between your ligand and protein residue.
***
## Submitting a Prediction
### Step 1: Create New Prediction
In Sandbox, click **New Prediction** to get started.
After naming your prediction with an identifiable description, add:
* Target protein
* Post-modified small molecule
* Any cofactors to be included in the complex
In this example, to model the bound acrylamide warhead, add the SMILES string for AMG 510 containing an ethyl-substituted amide.
![Setting up a covalent inhibitor prediction]()
### Step 2: Configure Bond Constraint
On the **Constraints** page, select **Bond** to guide the prediction. The Bond constraint is used to specify a covalent bond between different atoms in the complex.
1. **Select the Protein residue** that you wish the model to covalently modify
2. **Select the Ligand** from the left-hand side
For PDB ID 6OIM, choose:
* **Ligand**: MOV
* **Residue number**: 26
![Configuring bond constraints for covalent inhibitor]()
### Step 3: Define Atom Connections
Using the right-hand side drop-down menus and atom labeling schematic, select the atoms to be connected.
In this example:
* **Ligand atom**: C25
* **Protein residue atom**: SG
Click **Add Bond Constraint**, and continue through the remaining steps.
***
## Viewing Results
When inspecting the predicted structure, the covalent bond between ligand and protein is now present.
![Results view showing covalent bond formation]()
The prediction results will show the covalent bond formed between the selected atoms, along with standard Boltz prediction metrics including binding likelihood, predicted affinity, and structure confidence scores.
***
## Key Considerations
Ensure your input SMILES includes the appropriate warhead structure. For acrylamide-based inhibitors, use the ethyl-substituted amide form rather than the reactive acrylamide.
Common covalent modification sites include:
* Cysteine (SG atom)
* Serine (OG atom)
* Lysine (NZ atom)
Verify the target residue number matches your protein sequence.
Covalent inhibitor modeling is currently available only in Sandbox mode. Design Projects support for covalent inhibitors is planned for future releases.
# Example Contact: Molecular Glues
Source: https://docs.boltz.bio/user-guide/sandbox/Constraints/Example-Contact-with-Molecular-Glues
Guide to modeling molecular glue degraders and stabilizers in Design Projects
***
## Introduction
Molecular glues function by inducing or stabilizing protein-protein interactions, forming ternary complexes between a target protein, the glue molecule, and a binding partner (such as an E3 ligase or other effector protein).
Boltz Lab can facilitate modeling of these ternary complexes within the Sandbox and Design Projects by allowing you to specify multiple protein sequences.
To explore Boltz performance on your target, begin by adding two protein sequences in the Sandbox, and a molecular glue probe. Following the same steps as below for setting constraints (if needed) provides a good way to explore this target class, before beginning a Design Project.
This guide uses PDB entry 5HXB as a reference example of a molecular glue ternary complex.
***
## Creating a Molecular Glue Target
### Step 1: Set Up Design Project
When you create a **Target** within a Design Project, add **two protein sequences** instead of one:
1. **Protein 1**: Your target protein (e.g., degradation target)
2. **Protein 2**: Binding partner (e.g., E3 ligase, effector protein)
The platform will predict the ternary structure with both proteins present.
When you add your glue probe molecule, the model should predict a 'pocket' at the interface between the two proteins.
### Step 2: Define the Binding Site
There are two approaches to defining where the molecular glue binds:
For well-formed interface pockets like PDB 5HXB, the pocket prediction will automatically identify the binding site at the protein-protein interface.
This works when:
* The proteins have a clear interface cavity
* The glue binding site is well-defined
* Interface residues are pre-positioned for binding
For cases where the interface pocket is not automatically detected, manually select residues from **both proteins** on each side of the interface to define the binding space.
**How to select residues:**
1. Identify interface residues on Protein 1
2. Identify interface residues on Protein 2
3. Select residues flanking the expected glue binding site
4. This defines the space between the two proteins for pocket prediction
![Molecular glue binding at protein-protein interface]()
***
## Guiding Ternary Complex Formation
### Using Contact Constraints
If the predicted ternary complex does not match your expected orientation, use the **Contact Constraint** to guide the model.
The Contact constraint defines specific residue-residue interactions between the two proteins, helping to position them correctly relative to each other.
**When to use Contact constraints:**
* Proteins are predicted in the wrong relative orientation
* Known key interface contacts are not formed
* The glue binding site is not properly formed between proteins
Contact constraints work by specifying pairs of residues that should be in close proximity, helping the model to place the two proteins into the correct ternary complex geometry.
***
## Key Considerations
Choose appropriate protein constructs:
* Use domains that are known to interact (avoid including unstructured regions)
* Match experimental constructs when available
* Consider including known co-factors or modifications
The quality of interface pocket prediction depends on:
* Clarity of the protein-protein interface
* Presence of pre-existing cavity at the interface
* Degree of induced fit required
If predictions are poor, try adjusting which interface residues you select for pocket definition.
Ensure your input molecule represents the actual glue structure:
* Include all relevant functional groups
* Use appropriate protonation states
* Consider stereochemistry if relevant
Validate your predictions by:
* Comparing to known ternary complex structures (when available)
* Checking that predicted binding poses position the glue at the interface
* Verifying that key interface contacts are maintained
* Confirming predicted binding affinity is reasonable
# Overview
Source: https://docs.boltz.bio/user-guide/sandbox/Constraints/Overview
Guide predictions with structural knowledge to improve accuracy for challenging binding modes.
Constraints allow you to incorporate experimental knowledge or structural hypotheses directly into Boltz predictions. While standard predictions to well-defined binding sites work without constraints, they become essential for covalent attachments, cryptic pockets, and ambiguous binding modes.
For well-characterized binding interactions (such as reversible inhibitors at orthosteric sites, or known protein-protein interfaces), constraints are optional.
The model will find the binding site automatically.
***
## When Should You Use Constraints?
Specify specific atom-atom covalent bonds
Direct binders to non-obvious or allosteric binding sites
Enforce proximity between biomolecular chains or domains
Constraints are most valuable when you have prior structural knowledge—SAR data showing a covalent attachment, a crystal structure revealing an allosteric pocket, or biochemical evidence of a biomolecular interaction.
***
## Three Types of Constraints
### Bond Constraints
Define **covalent bonds** between specific atoms.
| Property | Description |
| ------------------- | ------------------------------------------------------------------------- |
| **Use Case** | Covalent bonds (acrylamide warheads, nitrile warheads, disulfide bridges) |
| **Inputs Required** | Chain + Residue + Atom for both bonding partners |
| **Model Behavior** | Hard constraint—these atoms MUST form a bond |
Bond constraints are essential for predicting structures with covalent attachments between any biomolecular components—such as covalent inhibitors (e.g., EGFR inhibitors targeting Cys797), disulfide bonds, or peptide cyclization.
Step-by-step guide for modeling covalent bonds
***
### Contact Constraints
Define **distance restraints** between atoms or residues.
| Property | Description |
| ------------------- | ---------------------------------------------------------------- |
| **Use Case** | Known interaction sites |
| **Inputs Required** | Two entities + maximum distance (4-20 Å) |
| **Model Behavior** | Predicts a structure using the given information as a constraint |
Contact constraints help when you know two regions should interact. They're useful for enforcing proximity between any pair of biomolecules—proteins, ligands, DNA, RNA, or combinations thereof (e.g., molecular glues bridging protein chains, or ligand-DNA contacts).
Guide multi-protein assemblies with contact restraints
***
### Pocket Constraints
Define **binding site residues** where a ligand/chain should bind.
| Property | Description |
| ------------------- | ---------------------------------------------------------- |
| **Use Case** | Cryptic pockets, allosteric sites, ambiguous binding modes |
| **Inputs Required** | Set of residues + which chain should bind |
| **Model Behavior** | Directs binder chain toward specified pocket region |
Pocket constraints are critical when your target has multiple potential binding sites and you want to focus on a specific one—such as an allosteric site distinct from the orthosteric pocket. The binder chain can also be a polymer, for example, to specify the epitope that an antibody binds to on the antigen.
Target specific binding regions with pocket constraints
***
## Using Constraints in Design Projects
Constraints become even more powerful in Design Projects, where they guide virtual screening campaigns and iterative design cycles. You can apply the same constraint logic to entire libraries of compounds, ensuring all predictions respect your structural requirements.
While constraints can also be used to test multiple hypotheses, pay attention: overly restrictive constraints can force the model into unrealistic conformations.
***
## What's Next?
Detailed instructions for configuring each constraint type
Apply constraints across virtual screening campaigns
# Setting Up Constraints
Source: https://docs.boltz.bio/user-guide/sandbox/Constraints/Setting-Up-Constraints
Step-by-step instructions for configuring bond, contact, and pocket constraints.
After uploading your molecules in Sandbox or Design Projects, navigate to the **Constraints** step. Here you can add structural guidance to improve prediction accuracy for challenging binding modes.
Constraints are optional - skip this step if your system doesn't require them.
For well-characterized binding interactions at defined sites, the model will automatically identify the binding site.
***
## Bond Constraints
Bond constraints define covalent bonds between specific atoms that are essential for modeling covalent attachments such as irreversible inhibitors (e.g., EGFR covalent drugs), peptide cyclization, or disulfide bonds.
### How to Set Up a Bond Constraint
Choose any two entities that will form the bond (protein chains, ligands, DNA, RNA) from the left panel. For example:
* **Protein A** and **Ligand D** for a covalent inhibitor
* **Protein A** and **Protein A** for a disulfide bond between residues
* **Protein A** and **DNA B** for a covalent DNA-protein crosslink
On the right side, use the dropdown menus to specify which atoms will form the covalent bond:
**Atom 1 (Protein side):**
* **Chain:** The protein chain letter (e.g., Chain A)
* **Residue:** The residue number and identity (e.g., 13C CYS)
* **Atom:** Select from available atoms (e.g., SG for cysteine sulfur)
**Atom 2 (Ligand side):**
* **Chain:** The ligand chain letter (e.g., Chain D)
* **Residue:** Ligand identifier (e.g., 94C)
* **Atom:** Select the reactive atom (e.g., C14 for acrylamide carbon)
The platform displays 2D structures to help you identify the correct atoms.
Click **Add Bond Constraint** to save. You can add multiple bond constraints if your system has more than one covalent attachment.
![Bond constraint setup interface]()
Bond constraints are **hard constraints**—the model must satisfy them. Only use bond constraints when you have strong evidence for a covalent interaction.
***
## Contact Constraints
Contact constraints define distance restraints between atoms or residues, ideal for known interaction sites.
### How to Set Up a Contact Constraint
Choose any two entities that should be in proximity. Contacts can be between any pair of biomolecules (protein, ligand, DNA, RNA), for example:
* **Protein-protein:** Two different residues
* **Protein-ligand:** A protein residue and a ligand
Example: Chain A residue 13C (CYS) and Chain B residue 120D (ASP) for a molecular glue bridging two proteins.
Depending on the entities selected, you may need to specify which atoms participate in the contact. For small molecules (ligands), select the specific atom. For polymer residues (protein, DNA, RNA), the constraint applies to one atom in the residue.
Use the slider to define the maximum allowed distance between the contact sites, ranging from **4Å to 20Å**:
* **4-6Å:** Tight interaction (hydrogen bonds, salt bridges)
* **8-12Å:** General proximity (nearby residues)
* **15-20Å:** Loose guidance (same binding region)
Toggle **"Enforce with potential"** to make the distance constraint stronger. This leverages the Boltz-steering technique to enforce the distance constraint.
Click **Add Contact Constraint** to save.
![Contact constraint setup interface]()
**Common Use Cases:**
* Molecular glues or bridging molecules requiring proximity between biomolecular chains
* Enforcing a known H-bond or salt bridge interaction
* Directing a binder toward a specific protein region without forcing exact geometry
***
## Pocket Constraints
Pocket constraints direct a binder to a specific region of the target protein by defining which residues form the binding site. This can be used for example to specify ligands binding to cryptic pockets, or allosteric sites, or specifying the epitope that an antibody should bind to on the antigen.
### How to Set Up a Pocket Constraint
From the **Binder Chain** dropdown, choose which entity should bind to the defined pocket (e.g., Chain B).
The binder chain itself cannot be used to define contact points. You're defining where the binder should go.
Click on residues in the protein sequence viewer to select the residues that form your target binding site. Selected residues will be highlighted in green.
The interface displays: **Chain A: 124, 204, 307, 317, 405** (example residues)
Use the slider to define how close the binder must be to the pocket residues (4-20Å). This sets the "binding site radius."
Toggle **"Enforce with potential"** to strengthen the pocket constraint. The model force itself to place the binder near the specified residues within the specified distance.
Click **Add Pocket Constraint** to save.
![Pocket constraint setup interface]()
**When to Use Pocket Constraints:**
* Targeting an allosteric site distinct from the orthosteric pocket
* Modeling binders to a cryptic pocket that's not obvious in the apo structure
* Disambiguating between multiple potential binding sites on the same protein
* Enforcing binding to a specific sub-pocket in a large binding cavity
***
## Tips for Effective Constraints
Add one constraint at a time and validate results before adding more. Over-constraining can force unrealistic structures.
Base constraints on SAR data, crystal structures, or biochemical assays.
Run a prediction without constraints as a baseline. If it fails to find the correct binding mode, then add constraints.
Check pLDDT and iPTM scores after adding constraints. Lower scores may indicate incompatible constraints.
***
## What's Next?
Walk through a real covalent warhead prediction
Model ternary complexes with contact constraints
Target allosteric or cryptic binding pockets
# Constraints in Small-Molecule Design Projects
Source: https://docs.boltz.bio/user-guide/sandbox/Constraints/Usage-in-Design-Projects
How to use constraints to refine predicted apo structures before virtual screening.
Constraints work differently in **Small-Molecule Design Projects** compared to Sandbox.
In Small-Molecule Design Projects, constraints are applied only to the **predicted apo structure** of your target protein - not to individual molecules in your virtual screening libraries.
**Key Information:** They only modify the target protein's unbound structure.
If you need to test specific binding hypotheses (e.g., "does this warhead form a covalent bond?"), use **Sandbox** with constraints for those individual predictions. Then incorporate successful molecules into your Design Project.
***
## The Design Projects Workflow
Understanding where constraints fit in the Small-Molecule Design Projects workflow is critical:
Add your protein target sequence via FASTA, UniProt, RCSB PDB, or file import.
Boltz predicts the unbound (apo) structure of your target protein automatically.
Review the predicted apo structure. If there are structural issues—incorrect domain orientation, problematic loops, or conformational problems - you can add constraints to guide the model toward a better structure.
**This is the only step where constraints are used in Design Projects.**
Select residues that form your target binding site. Optionally provide probe molecules to help guide pocket prediction.
Screen your molecule libraries against the refined target structure. Individual molecules in these libraries cannot have constraints applied.
***
## When to Use Constraints in Design Projects
Constraints in Small-Molecule Design Projects serve a **quality control** function - fixing structural predictions before virtual screening begins. Use them when:
Multi-domain proteins predicted in non-native orientations that would block the intended binding site
Critical loops (e.g., DFG motif in kinases) predicted in inactive conformations when you need the active state
Allosteric or cryptic pockets that require specific residue positioning to remain accessible
Forcing a particular conformational state (open/closed, active/inactive) relevant to your screening hypothesis
**"Only use this target if the structure is predicted well."**
***
## How to Apply Constraints to Apo Structures
During the **"Verify Unbound Structure"** step, you'll see the predicted apo structure with the option to add constraints.
![Design Projects constraint workflow]()
### Adding Constraints
Click **"+ Add Constraint"** on the right panel to open the constraint editor. All three constraint types are available:
| Constraint Type | Use Case in Apo Structure Refinement |
| --------------- | ------------------------------------------------------------------------------------------ |
| **Bond** | Force disulfide bonds or covalent modifications not predicted correctly |
| **Contact** | Enforce proximity between domains or secondary structure elements |
| **Pocket** | Define where the binding pocket should form (alternative to manual pocket selection later) |
**Contact constraints** are most commonly used in Design Projects to fix domain-domain interfaces or enforce known structural features from experimental data (crystal structures, cryo-EM).
The constraint setup process is identical to Sandbox - refer to the [Setting Up Constraints](/user-guide/sandbox/Constraints/Setting-Up-Constraints) guide for detailed instructions.
***
## Best Practices for Design Projects
Review the apo structure without constraints first. Only add them if you see clear structural problems.
Base constraints on crystal structures, homology models, or known domain interfaces.
If unsure whether constraints will help, run a quick Sandbox prediction with and without constraints to compare.
After applying constraints and predicting the pocket structure, verify that the pocket is positioned correctly. Poor constraints can shift the pocket to the wrong location.
***
## Sandbox vs. Design Projects: When to Use Each
| Scenario | Recommended Workflow |
| -------------------------------------------------------- | ------------------------------------------------------------- |
| Testing a specific covalent warhead hypothesis | **Sandbox** with bond constraints |
| Screening 50,000 molecules against a kinase | **Design Projects** (no molecular constraints) |
| Fixing multi-domain protein orientation before screening | **Design Projects** with contact constraints on apo structure |
| Validating a molecular glue ternary complex | **Sandbox** with contact constraints |
| Targeting a cryptic allosteric pocket | **Design Projects** with pocket constraints on apo structure |
| Exploring different binding modes of one molecule | **Sandbox** without constraints (let model explore) |
***
## What's Next?
Detailed instructions for configuring bond, contact, and pocket constraints
Learn more about the Design Projects workflow
Complete guide to creating and refining target structures
Return to the Constraints overview for Sandbox workflows
# Sandbox
Source: https://docs.boltz.bio/user-guide/sandbox/Overview
Your personal workspace for rapid Boltz predictions.
The Sandbox is the best place to start running your first predictions in Boltz Lab.
Here, you can quickly test hypotheses, validate binding poses, explore protein structures - all in a private workspace where you can experiment freely before committing to larger screening campaigns.
![Image]()
***
## Why Use Sandbox?
Test a binding hypothesis in minutes
Iterate quickly setting different constraints
Your predictions are visible only to you
***
## What Can You Predict?
Sandbox supports the prediction of the structure of arbitrary biomolecular complexes. Common use cases include:
| System Type | Example Use Cases |
| -------------------- | --------------------------------------------------------------------- |
| **Protein-Ligand** | Validate docking poses, predict binding modes, assess ligand affinity |
| **Protein-Protein** | Model complex interfaces, predict interaction sites |
| **Protein-RNA/DNA** | Explore nucleic acid binding proteins |
| **Apo Structures** | Predict protein conformations |
| **Covalent Binders** | Model irreversible inhibitors with bond constraints |
Sandbox is ideal for **one-off predictions** - validating a single compound, checking a crystal structure hypothesis, or exploring a new target before setting up a full Design Project.
***
## Your Prediction History
![Image]()
All your Sandbox predictions are stored and searchable. Use the table view to:
* **Search** predictions by name
* **Filter** by status or date
* **Sort** columns to find what you need
* **Duplicate** previous predictions as starting points
![Image]()
Hover over any row to display the three-dot menu for quick actions:
* **Duplicate prediction** - Clone all inputs to start a new prediction
* **Download results** - Export structures and data for offline analysis
***
## What's Next?
Step-by-step guide to creating predictions
Interpret confidence scores, view structures, and analyze binding
# Submitting a Prediction
Source: https://docs.boltz.bio/user-guide/sandbox/Submitting-a-Prediction
How to submit a prediction in Sandbox mode
In **Sandbox**, click **New Prediction** to get started.
![Sandbox New]()
After naming your prediction with an identifiable description, add polymers, small molecules and cofactors to be included in the complex.
![Image]()
## Adding Components
Import a .cif file or provide an RCSB PDB code to load all protein, RNA and DNA sequences, ligands and cofactors present in the file.
Import small molecules using a SMILES string, the structure drawer, or from an RCSB CCD code.
Add protein, RNA and DNA sequences individually from UniProt or FASTA strings. Modified amino acids and nucleic acids can be specified when adding a sequence.
***
## Constraints
On the **Constraints** page, add any bond, contact or pocket-contact information to guide the prediction.
Alternatively, skip this step to continue without constraints.
| Constraint Type | Description |
| :-------------- | :------------------------------------------------------------------------------------------------------- |
| **Bond** | Specify a covalent bond between atoms in the complex. Important for covalent ligands or cyclic proteins. |
| **Contact** | Specify a maximum distance (4-20 Å) between a pair of atoms as a hint to the model. |
| **Pocket** | Specify residues that a particular binder chain must be within a certain distance from. |
**Please see our dedicated Constraints documentation for full details**
***
## Affinity Prediction
The **Affinity** page lets you select a ligand to calculate binding affinity for ligand-protein interactions, or skip this step to run a structure-only prediction.
![Image]()
Only ligands with exactly one copy are allowed for affinity calculation.
When you're ready, click **Submit Prediction**.
# Viewing Results
Source: https://docs.boltz.bio/user-guide/sandbox/Viewing-Results
How to review results in Sandbox mode.
Once predictions are complete, the database is populated with the outputs.
![Image]()
Clicking a row opens the results viewer window.
To rename a prediction, select the edit button next to the prediction title name.
The right-hand side information boxes provide:
* **Boltz Prediction Properties** - Model scoring metrics including affinity metrics (if selected)
* **RDKit Descriptors** - Standard molecular property calculations
![Image]()
The 3D viewer displays the structure colored by per-residue pLDDT. Hovering over a residue, ligand, or cofactor in the viewer (or within the sequence window below) displays additional per-residue information.
***
## Interaction Analysis
Clicking a ligand or molecule in the viewer reveals additional interactions.
![Download]()
For more information on using the viewer, see the [Mol\* (MolStar) user guide](https://molstar.org/viewer-docs/).
If you wish to view or process in alternative software, the **Download** button allows export of the structural prediction and associated data for offline use.