David Minh

Looking at pipeline.py (https://github.com/choderalab/yank/blob/972753ad0f1e3b3b332e2a2b10a7ff6808f0474d/Yank/pipeline.py#L1074) it looks like if I have the correct names for the prmtop and inpcrd files then YANK will bypass its own setup.

Even if the new force fields are not much more accurate than GAFF, they have more chemical space coverage. I was trying to set up some calculations of [several known...

I ran it again with the same problem but [let the error logs complete](https://github.com/daveminh/Chem456/tree/master/static_files/tutorials/3cl-pro/YANK/output/ZINC000001714738/experiments/nan-error-logs).

Some of the systems failed with GAFF. But I can try to get one of them working...

Yeah, I’m working on the failures. I was trying to make a pipeline around it by building the inpcrd and prmtop files directly. > On Mar 23, 2020, at 9:09...

Just tried it and it works for me. Try again? https://github.com/daveminh/Chem456/tree/master/static_files/tutorials/3cl-pro/YANK/output/ZINC000001714738/experiments/nan-error-logs > On Mar 23, 2020, at 9:15 PM, John Chodera wrote: > > > The log link you posted...

The prmtop are in ../0-build/ > On Mar 23, 2020, at 9:46 PM, John Chodera wrote: > > > Do you have the prmtop files in one of your branches?...

The paper also identifies cinanserin as a weaker binder. That’s helpful for testing dynamic range since the dataset would have three orders of magnitude. > On Mar 23, 2020, at...

GAFF setup seems to work for ZINC000001714738, ZINC000002015152, and ZINC000003951740. Antechamber complains about carbon valences in Tideglusib (ZINC000013985228) and Carmofur (ZINC000001542916).

[This is a YAML that doesn't work with antechamber](https://github.com/daveminh/Chem456/blob/master/static_files/tutorials/3cl-pro/YANK/yaml/MPro_ZINC000013985228.yaml).