Estimation of mobile elements insertions frequencies in heterogeneous samples
Entering edit mode
8 weeks ago


I need to estimate frequency of mobile elements rearrangements in heterogenous sample (Illumina PE150). Specifically: bacterial sample went through experimental evolution process and the sample was sequenced. Therefore it represents a mixture of different subpopulations. I need a tool that will say that there is a mobile element inserted in some locus in X% of population (e.g. 20%). Or a tool that will show that there is an insertion and output a DNA sequence that later I would be able to identify as mobile element.

It sounds like a somatic variant calling in cancer genomics where you need to estimate what is going on in the heterogenous tumor. And it seems that it is an important task that should be solved. But I can't find tools for this task.

Does anyone use such tools?

P.S. I know about breseq - but it is not suitable for my samples for some reasons. Also I developed my tool iJump but I would like to see if there is anything else available.


heterogeneous reads calling elements illumina mobile varian short • 258 views
Entering edit mode

I think mobile element insertion callers don't have ploidy and allele balance assumption in their models. Is there a tool that will estimate the percentage of the mixture which has this mobile element insertion? It depends on the evolutionary model. There are specific tumor callers which assume some evolutionary models (e.g. fast clonal expansions with large reproductive benefits) and it may be suboptimal for your bacteria.

In cancer there are also copy-number changes which allow estimation of % content of different clones (and then each variant is assigned to the most likely clone). Point mutations are quite less informative for clonal decomposition.

Entering edit mode

Thank you for an answer. It matches my experience. I either did not see mobile element insertion callers that would assume mixed samples. Although clonal decomposition would be ideal for my task just percentage of affected population should be enough.

I've tried some clonal decomposition algorithms. But they failed for my bacterial samples likely as evolutionary assumptions do not work. At least in my experiments copy number variations in bacteria much more rare event than in cancer to be used in clonal decomposition.


Login before adding your answer.

Traffic: 1845 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6