AlphaFold 3 is now available on the SCC. This software proivdes an AI model developed by Google DeepMind and Isomorphic Labs for generating 3D predictions of biological molecules. To see the available versions, use the module command in an SCC terminal window:

[rcs@scc1 ~] module avail alphafold3

Model Parameters Terms of Use

Google DeepMind has changed the license structure of the Alphafold software for the version 3 release. The software source code remains open source. The model parameters are the result of training the AlphaFold model and are required to run the inference calculation that predicts molecular structure. The model parameters are distributed separately from the source code with their own terms of use.  SCC users who wish to run AlphaFold 3 need to read and understand the terms of use for the model parameters:

Obtaining the Model Parameters

You can request a copy of the model parameters by filling out and submitting this form. Approval of the request is entirely at the discretion of Google DeepMind. RCS staff cannot assist with filling out the form or intervene on your behalf with Google DeepMind if the request is rejected.

Note:In order to comply with the terms of use, each AlphaFold 3 user on the SCC needs to request, download, and use their own personal copy of the model parameters.

Here are instructions for filling out the form:

  • Read and understand the section titled Key things to know when using the AlphaFold 3 model parameters and output.
  • In the first email field (titled “Email”), enter your BU email address.
  • In the second email field (titled “Google account email address”) you must enter an email address that ends in gmail.com if you are faculty or staff who uses Outlook Mail.  If you use your BU email address in this field the request to access the model parameters will be rejected. If you don’t have a Gmail email address, visit gmail.com and create one. It has been reported to RCS that student email addresses that end in bu.edu work here as the student email system uses Gmail.
  • In the field titled “URL of public-facing website for non-commercial organization“, enter https://www.bu.edu
  • When all the fields on the first page are filled in, click the Next button.
  • The second page has a single question that asks “Do you intend to provide access to the AlphaFold 3 model parameters to other researchers within your non-commercial organization? (E.g. as part of a centrally managed computing cluster)“.  Answer No.
  • Click Next, read the text on the last page, and if you are satisfied you understand the terms of use, complete the form and submit it.

The waiting time for a reply from the form submission ranges from a few hours to a few days.

Download the Model Parameters

An email will be sent to the email address provided in the first email field on the request form with a download link. Clicking the link will connect you to a Google Drive page where you can download the model parameters. The parameters consist of a single file approximately 1GB in size. This file can then be stored on the SCC in your home directory or in your project’s /projectnb or /rprojectnb directory.

Running AlphaFold 3

The AlphaFold model prediction consists of two steps: a data pipeline and database search that runs on CPUs only (without a GPU), and the model inference which requires a GPU. The data pipeline step is memory and CPU intensive and can take many hours to run. The inference step requires a GPU and, in RCS testing, runs significantly faster than the previous step. The best practice on the SCC is to submit an AlphaFold calculation as two separate jobs – a CPU data pipeline job that uses at least 8 cores, followed by a GPU inference job.

For instructions on running the software consult the documentation that is available within the module using the help command:

[rcs@scc1 ~] module help alphafold3/3.0.0