ncbi / AF2_benchmark

12 stars 3 forks source link

Scripts for generating plots and figures for AlphaFold predictions of fold-switched conformations are driven by structure memorization

MSA and Predictions generated using AlphaFold2, ColabFold, MSA clustering done using AF_Cluster

TM-align was used to calculate TM-score, using python binding around the TM-align code for structural alignment of proteins

check these links for more details:

1. https://github.com/google-deepmind/alphafold

2. https://github.com/sokrypton/ColabFold

3. https://github.com/HWaymentSteele/AF_Cluster/tree/main

4. https://zhanggroup.org/TM-align/ and https://pypi.org/project/tmtools/

Packages needed in python : Matplotlib, Seaborn, Pandas, Numpy, tmtools

*Important* please change the paths in the python scripts before running them

MSAs along with related log, output files generated from AF_Cluster, are in AFcluster_MSAs

Code was run and tested on LINUX and MAC OSX machines.