Welcome to MULTICOM_ligand’s documentation!¶
Comprehensive ensembling of protein-ligand structure and affinity prediction methods
Contents:
- Installation
- DynamicBind checkpoint (~0.25 GB)
- Tutorials
- RoseTTAFold-All-Atom checkpoint (~1.5 GB)
- How to prepare MULTICOM_ligand data
- Available inference methods
- How to run inference with individual methods
- How to run inference with a method ensemble
- NOTE: the suffixes for both
output_dirandensemble_benchmarking_repeat_indexshould be modified to e.g., 2, 3, … - benchmark using the PoseBusters Benchmark dataset e.g., after generating 40 complexes per target with each method
- benchmark using the Astex Diverse dataset e.g., after generating 40 complexes per target with each method
- benchmark using the DockGen dataset e.g., after generating 40 complexes per target with each method
- benchmark using the CASP15 dataset e.g., after generating 40 complexes per target with each method
- analyze benchmarking results for the PoseBusters Benchmark dataset
- How to create comparative plots of inference results
- analyze benchmarking results for the CASP15 dataset
- For developers
- Acknowledgements
- Citing this work
- Bonus