search

brentp / duphold

don't get DUP'ed or DEL'ed by your putative SVs.
MIT License
101 stars 9 forks source link

local fold-change #5

Closed brentp closed 6 years ago

brentp commented 6 years ago

currently there is a global fold-change and a GC fold-change.

currently, if a deletion falls into a relatively high-coverage area it may not appear to have any coverage loss (FC~1) but it could be lower than the surrounding bases. This is what the eye sees when looking at SVs.

this would add a local fold change. it would look left and right of the event by ~500 bases and determine if the inside of the event differed from the outside.

this would help to annotate events that have fall into regions (e.g. megabase) that are largely different than the rest of the genome.

brentp commented 6 years ago

there is a nice test-case for this in syndip at 1:144,952,651-144,956,575.

local-fc

since the local coverage is higher, the duphold DHBFC for this is 1.11 but there would be a clear drop if we compared to the flanks. could be DHFFC for flank-fold-change. Number=2 (1 for left, 1 for right).