ACES: AlphaFold Protein Structure Prediction
Overview
Instructor(s): Dr. Michael Dickens
Time: Tuesday, October 7, 2025 — 10:00AM-12:30PM CT
Location: Online using Zoom
Prerequisite(s): Current ACCESS ID, basic Linux/Unix skills
This short course covers how to run AlphaFold on the ACES cluster, utilizing the NVIDIA H100 GPUs.
A Registration button will appear here when registration has been opened.
Course Materials
Learning Objectives and Agenda
In this class, participants will:
- Learn about AlphaFold resources and Limitations
- Learn about shared database files
- Learn how to run ParaFold to reduce GPU idle time
- Submit an example ParaFold/AlphaFold2 job script
- Submit an example AlphaFold3 job script
- Visualize the AlphaFold job results
This course focuses, among others, on the following topics:
- AlphaFold History
- Running AlphaFold2 on ACES
- ParaFold AlphaFold2 workflow
- ParaFold with reduced_dbs
- AlphaFold2 confidence metrics
- Running AlphaFold3 on ACES
- HPRC Cluster Utilities
- Visualization of Results
- view predictions in Jmol
- Job Resource Monitoring
Note: In order to run AlphaFold3 on the HPRC ACES cluster, you will need to request the access model parameters file by agreeing to the Google terms of service
Note: During the class sessions many aspects of the material will be illustrated live via a login to ACES. Attendees will follow along and complete the exercises on their own computer.
You are encouraged to contact the HPRC helpdesk with any questions regarding ACES.