Date: May 5th, 2022
Attendees: Bharath, Ashish,
Bharath has been working to review GSoC applications for this cycle. Bharath has also been working on some scaled DeepChem deployments and has some ideas for scaling that might make it into DeepChem.
Ashish looked at the counting cells tutorial and the MoleculeNet tutorial.
Date: May 6th, 2022
Attendees: Bharath, Shai, Arun, Stanley, Peter, Abhishek
Bharath gave the same update as at the previous call.
Shai has been working on schoolwork primarily. Shai looked into free AWS credits and asked about the requirements for credit cards from these programs.
Arun has been working on his PR https://github.com/deepchem/deepchem/pull/2903 addressing comments.
Stanley has been working on protein transformers and is working with David and Paulina to gather some interesting datasets and ideas.
Peter has been reviewing GSoC proposals this week.
Abhishek has been reviewing GSoC proposals.
DeepChem Bioinformatics Call
Date: May 9th, 2022
Attendees: Tony Davis, Ashish Makani, Arun Palaniappan, Stanley Bishop, Bharath Ramsundar, Paulina Paiz
Bharath has been reading a couple of papers. CLoud training infrastructure that would be useful for this group.
Ashish first time in this call. Works on drug development in oncology.
Looking into integrating cell packa
Tony has been implementing secondary structure prediction. RaptorX software trying to see if there is something more Python-friendly. Looking at other repos that do secondary structure prediction and then plug it in to DeepChem.
Stanley has been meeting with David about the design of the attention mechanism and what it can learn about motifs in a sequence. Starting point: proteomic datasets for unit testing and engineering verification for the protein transformers. PR for DeepBio basic load up for hugging face transformers to get comparisons.
Paulina has been looking into setting up citations for the colab tutorials. Stanley and Bharath suggested some methods for updating on DeepChem. Paulina has been looking into scvi-tools and mentioned that there were some errors in earlier versions of scvi-tools that have been fixed in later versions.
Think about year tutorial was edited.
Renamed most of the tutorials.
Bharath recommends alphabetical listing and using the blame functionality on Github.
Ashish has been looking at ilastik and cell profiler as potential tools for writing a tutorial on cell counting for the deepchem sequence. Paulina also suggested looking at the watershed algorithm.
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.