This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Identifying CNVs from targeted amplicon sequencing

I have few unmatched samples which have been sequenced with targeted amplicon sequencing using a custom library which covers a small segment of a chromosome (less than 300kb in length). The samples have been aligned to a BAM file and now I have to check for copy number variants.

I've used CNVkit to detect CN regions, however the docs mention that it's better if the sequenced regions span greater than million bases in length. I've also used ONCOCNV but it fails to detect any CN regions (although visually the shift in CN is visible).

Since only small region of the samples are covered, are there any other tools I can use, or do I have to filter the CN regions manually? Are there any papers which can help me out on how to manually filter out CN regions for the small segments?

ngs amplicon sequencing sequence

I know that some people at my institute are using HMMcopy. I am not sure of its performance on targeted NGS. There are so many NGS copy number tools such that it's difficult to know which ones are best. The earlier ones that were released typically did not adjust for GC content, which renders their utility as minimal.

3 answers

I have been collecting some papers about CNV analysis (link). There is a topic about Amplicon sequencing (AS). Nowadays, I'm using Exomedepth+cnvScan. I tested CoNVaDING, but didn't have good results with my data. Now I'm testing VisCap.

Cheers.

Dear geocarvalho,

Do you still use Viscap ?

Thanks

I didn't finish my benchmark, but I helped a friend to use it!

I have used CONTRA (https://www.ncbi.nlm.nih.gov/pubmed/22474122). This software is supposedly tuned for targeted sequencing.

you can try CONVector, it is designed for germinal CNV detection in the PCR-enriched targeted sequencing data. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5075217/pdf/12859_2016_Article_1272.pdf

Please create a new Tool type post and showcase your tool. Bumping old posts with the same content comes across as spamming and is not encouraged.

Log in to answer this question.