Question: SomaticSignatures: extract low level information
0
3.1 years ago by
mathieu.bourgey0 wrote:

Hello,

first of all thank for this very nice package.

I generate the somatic mutation decomposition of my data using a meta data information to group my samples (specified using the argument group of the motifMatrix function).

Everything works well and I'm able to get the signature decomposition and the contribution of each of these signatures to my different groups.

What I would like to know is if it possible then to have the contribution of each of these signatures to my whole set of samples ?

If I tried to estimate the signature of my sample directly with the same number of expected signatures but the new signatures are different from those I found using my groups.

do yo have any idea ?

Thank you

Mathieu

modified 3.1 years ago by Julian Gehring1.3k • written 3.1 years ago by mathieu.bourgey0
Answer: SomaticSignatures: extract low level information
0
3.1 years ago by
Julian Gehring1.3k
Julian Gehring1.3k wrote:

You can get the total contribution of each signature to all your samples by summing up the contributions over all samples/groups. Here an example with data from the package:

data("sca_mm", package = "SomaticSignatures")
sigs = identifySignatures(sca_mm, 5)

samples(sigs) ## contribution of each signature (columns) to each sample (rows)
total_contribution = colSums(samples(sigs)) ## sum over samples (rows)
total_contribution