search

Gallicchio-Lab / AToM-OpenMM

OpenMM-based framework for absolute and relative binding free energy calculations with the Alchemical Transfer Method
Other
110 stars 31 forks source link

Positive ABFE result #90

Closed nonsensejoke closed 1 month ago

nonsensejoke commented 1 month ago

Hello Professor Gallicchio,

I tried to run ABFE simulation with AToM for a protein+ligand system built with my own tool chain. I first obtained AMBER prm7/rst7 files, then convert them into xml/pdb files for OpenMM.

I tried several ligands against the same receptor protein and found that all my ABFE results are positive (I expect negative values as the ABFE examples). I wonder what the cause can be in this situation.

In the zip file below, I have attached the cntl file, uwham_analysis input/output, the analyze.sh script and Rplots.pdf. atom-abfe-result.zip

Thank you for your time.

egallicc commented 1 month ago

The ligand is probably too large (56 atoms) for ABFE. The perturbation energy distributions (see Rplots.pdf) are disconnected. I recommend binding a fragment of the ligand first and then use RBFE to progressively build the full ligand. See https://pubs.rsc.org/en/content/articlelanding/2023/cp/d3cp02125d for an example.

I also recommend double-checking the location of the binding site region (the LIGAND_CM_ATOMS, RCPT_CM_ATOMS, CM_KF, and CM_TOL settings) to confirm that it is consistent with the desired binding pose.

nonsensejoke commented 1 month ago

Thank you very much for your suggestion and the paper. I will look into this.