Data Sources - Citations?

Mariewelt / OpenChem

OpenChem: Deep Learning toolkit for Computational Chemistry and Drug Design Research

https://mariewelt.github.io/OpenChem/

MIT License

681 stars 114 forks source link

Data Sources - Citations? #24

Open AcylSilane opened 3 years ago

AcylSilane commented 3 years ago

I am curious - what are the sources for the benchmark datasets that are supplied with OpenChem (e.g. the logP, melting temperature, etc datasets)? I checked the documentation and readme, but wasn't able to find anything.

Thanks!

rgerkin commented 2 years ago

I am also curious about this. Is it experimental data? Is it generated by prior models?

isayev commented 2 years ago

Ups, sorry for missing this. @AcylSilane @rgerkin: Those are manually curated (removed errors, duplicates, etc.) datasets of experimental properties. LogP is obtained mostly from the PHYSPROP database. Melting data is from a couple of other publications. We describe them in the OpenChem paper https://doi.org/10.1021/acs.jcim.0c00971