Mutation Sets

Mutation datasets

• Humsavar : Deleterious mutations and polymorphisms currated by UniProt. Contains mutations associated with different heritable diseases and cancers, as well as polymorphisms.

• ClinVar : Deleterious mutations and polymorphisms currated by UniProt.

• COSMIC : Mutations that have been found by sequencing cancer cells. Not clear which mutations are "drivers" and which mutations are passengers.

• DoCM : Database of Curated Mutations. Contains currated mutations which are known to be drivers in cancer.

Small-scale studies

• CAGI4 SUMO ligase : Mutatons affecting the activity of human SUMO ligase (UBE2l), measured using a high-throughput yeast complementation assay.

Notes:

• A good application for semi-supervised learning would be to train on COSMIC + DoCM.