tetedange13
commented
3 years ago Hi @zhuhenan,
Thanks for sharing your files, but it can be hard to explain your results without knowing how you generated them (exact CNVkit commands you used)
I will assume you have WES data and in a case with both normal and tumor samples
=> When CNVkit is used this way, fix step will calculate a log_ratio value per region/bin, by comparing coverage in current tumor sample ("antitargetcoverage.cnn", "targetcoverage.cnn") with coverage on same region from a reference produced with all "normal" samples you gave ("reference.cnn")
=> Following segment step will use these log_ratio values to determine boundaries of segments with same log_ratio and their corresponding height
So looking at an individual value of coverage of some sample will not actually give you a lot of information
=> If you have for this region a log_ratio=-1.4 and coverage=~300X , it means there you have several bins where 300X can be considered a lower coverage regarding you reference coverage there
=> I suggest you to look also at your "reference.cnn" file, how are covered these regions overlapping your deleted segment?
Also if you visualise your cnr,cns files with scatter (exact parameters: --by-bin -c chr16:2089716-3136227 -s chr16.rename.cns chr16.rename.cnr)
=> You see this deleted orange segment on PDK1 looks justified, based on associated gray probes position (lower in this genomic area):
I hope this answered your question. Have a nice day. Felix.
zhuhenan
Hi,
I got confused about the deletion of the PKD1 gene found in some of my samples. The cns files show a breakpoint in the PKD1 gene suggesting a completed deletion of 1/3 gene. But when I check the cnn and cnr files, the coverage of the PKD1 is ~300x. So I am not sure why the pipeline suggests a deletion in this region.
I have attached the chr16 of cnn, cnr, and cns files from one of my samples.
Best, Henan
chr16.rename.zip