DeepChem Minutes
India/Asia/Pacific Call
Date: February 4th, 2021
Attendees: Bharath, Peter, Yuanqi
Summary:
Bharath has been busy and couldn’t do much.
Peter has been busy and couldn’t do much.
Yuanqi sent in a PR to add SIDER to moleculenet. Yuanqi asked if we should add more models to MoleculeNet and Bharath said we definitely should now that we have more infrastructure to benchmark models.
Americas/Europe/Africa/Middle East
India/Asia/Pacific Call
Date: February 5th, 2021
Attendees: Bharath, David, Vignesh, Maria, Seyone, Nathan
Summary: We had a couple new attendees so we started with intros.
Maksym has a background in drug discovery, a year at Harvard, and a couple years at MIT, and is a PhD student with Yoshua Bengio where they work to apply reinforcementlearning to find drugs.
Seyone has been working on benchmarking the smiles tokenizer models on different models, and has been working to compare the differentiation in tokenization between the two models to understand the boost there. Seyone has also been working on a talk alongside Bharath for Carnegie Mellon University’s SciML webinar.
Nathan this week finished up a PR to add Gnina as a backend to a molecular docking API and should be ready to merge in soon.
David has been working on migrating code over from mpi to rise to pyscreener and will plan to start a pull request
Maria has been continuing to work on the PR for fragment featurization into ConvMolFeaturizer. Maria has been discussing with Peter on Github to make adjustments to the PR. Maria is thinking right now that the best place to make the change would be to write a special transformer that will flatten a dataset of fragments.
Vignesh has been working on the LCNNModel and is waiting for reviews and is now working on a PAGTNModel
Joining the DeepChem Developer Calls
As a quick reminder to anyone reading along, the DeepChem developer calls are open to the public! If you’re interested in attending either or both of the calls, please send an email to X.Y@gmail.com, where X=bharath, Y=ramsundar.