Lincoln Harris
CZ Biohub, January, 2019
I have a feeling that if i calculate the relationship between number of mutations and expression/coverage, something good will come of it.
- downsample dataset
maybe all of the cells from 2 given clusters - calculate coverage, transcriptome wide
(bedtools?)
how much of this info do we already have, in vcf files? - select gene panel of interest
start with EGFR and Actin, maybe - normalize expression/coverage across all samples, for GOI
(FPKM?) - plot number mutations v. normalized expression
grab mutation info from
makeGeneCounts_parallel.py
output (s3://darmanis-group/singlecell_lungadeno/non_immune/analysis/geneCellMutationCounts.csv)