Irfan Alibay

Link or store "known" computed results?

3

Especially if this ends up being a per-release benchmark run, it would be good to have a simple manner for folks to work out what historical computed free energies were...

Presence of two PDBs for Thrombin

2

Adding this here so we can track progress. There are two PDBs in the `crd` folder for thrombin (protein.pdb and thrombin_prepped.pdb), we need to work out which one needs keeping.

Ahead of 0.3.0, we need to have some kind of documentation that specific outlines the changes and the process taken to create this new version of the benchmark set. I...

Ideally we should only have the ligands.sdf file, are they safe to remove?

Investigate clashes

2

The following clashes were identified by @dotsdl ``` cmet: - lig_CHEMBL3402745_200_5 - lig_CHEMBL3402745_200_5 hif2a: - lig_224 ```

Several issues all linked together here. Tests are currently failing due to issues with the API, and CI is running on some old "example" files which need to be removed.

Generate benchmark data

1

Opening a separate issue so we don't lose all the information when #52 gets merged. Based on discussions the following needs to happen before we move ahead with benchmarking properly:...

Store assay specific information in `targets.yaml`

4

Following from @bobym's very detailed dive into how these systems were actually prepared, it has become very apparent that we need to expand the details in `targets.yaml` to capture this...

There's been quite a few discussions about this over various different platforms so we decided it would be best to have a single place to aggregate discussions. The short story...

Identified by @bobym The dois for the measurements in https://github.com/openforcefield/protein-ligand-benchmark/blob/main/data/2020-06-19_eg5/00_data/ligands.yml point to the hif2a paper rather than the one for EG5.