https://github.com/dphansti/CORAL

Do not hardcode priority of data sources, it'd be something like:

Kinases,KinHub,KLIFS,PKinFam
Kinase A,True,True,True
Kinase B,False,True,False
Kinase C,False,False,True
....

List of small things that could be added to the table

• compound id: allows to refind or compare compounds (not just smiles)
• max phase: would allow us to generate our own kinase inhibitor set and host it here

We need to check how well ChEMBL annotates mutations. Mutant information can be found in the columns Assay Variant Mutation and Assay Variant Accession.

Example data point:

https://www.ebi.ac.uk/chembl/g/#browse/activities/filter/molecule_chembl_id%3A(%22CHEMBL3354189%22)%20AND%20standard_type%3A(%22Ki%22)

Over the past few days kinase.com was not reachable. That site is used to cross-reference human kinases in human_kinases.ipynb.

Some kinases might be missing.

A new version of ChEMBL is out. Should the zip file be adapted to v.29?

• Webserver: http://dunbrack3.fccc.edu/kincore
• Static tsv: http://dunbrack3.fccc.edu/kincore/static/downloads/text-files/Human_Allgroups_Allspatials_Alldihedrals_All.tsv

Much of the necessary code is merged.

The release draft is here.

Remaining tasks

Code

• update the EGFR notebook (references to the current release; the last release has the non-updated version...)
• add data/human_kinases_and_chembl_targets.chembl_33.csv
• Both changes are done with merging #20

Release artifacts

• package the code as kinodata-0.5
• create the activities artifacts
• create the EGFR activities artifacts

Release notes

• get the numbers for the table in the release post
• mention removal of kinase.com
• mention changed Uniprot query

The summary data in the last release (https://github.com/openkinome/kinodata/releases/tag/v0.2) differs from what reported by the notebooks and in the data files that are outputted by them:

vs the number of unique records in the output .csv's and reported in the notebooks:

The notebooks appear to run fine so I have added the data from the notebooks/output files themselves to the latest release (https://github.com/openkinome/kinodata/releases/tag/v0.3). But having looked at the data directly, I can't establish where the data in the first table comes from.