Investigating the conformational landscape of AlphaFold2-predicted protein kinase structures

All scripts used to generate the figures in the paper "Investigating the conformational landscape of AlphaFold2-predicted protein kinase structures" by Al-Masri et al. 2023 are included.

All inputs needed are in the inputs directory and all outputs generated in outputs. The scripts include:

AF2 Database Structure Conformations & Pocket Residue Fingerprints

1. AlphaFold_fingerprints.ipynb

• Extracts the pocket residues of AF2 structures
• Generates KiSSim fingerprints for pocket residues
• Performs t-SNE on distance fingerprints

2. plot_conformations.ipynb

• For both AF2 and PDB structures, the distribution of conformations by species & kinase group are plotted

3. plot_plddt.ipynb

• Plots average pLDDt for all pocket residues

Requirements

Python # 3.6.8+ biopython # 1.77+ pandas # 1.4+ numpy # 1.22+ plotly # 5.8+ sklearn # 1.0.2+ kissim # 1.0.0

Docking and MD simulations

4. plot_rmsd.py, plot_rmsf.py, and plot_MD_interaction_anaysis.py

• These script were used to create the plots to analyze the MD simulations

5. plot_docking_enrichment.py

• This file contains the functions used to compute and plot the enrichment from docking.

6. crossdocking_interaction_analysis.ipynb

• This notebook contains the functions used to create the plot the interactions analysis from docking.

Requirements

Python # 3.7.12+ numpy # 1.21.6 pandas # 1.4.2 matplotlib # 3.5.2 seaborn # 0.11.2 MDAnalysis # 2.1.0 prolif # 2.0.0 biopython # 1.79 mdtraj # 1.9.6