WGS CNV data input

[nanshan yang]

Dear  Joseph Glessner:    

    HI

    I use WGS data for CNV analysis and the software is CNVkit.The result of the CNVkit file only have  deletion and duplication

    

    could you please give me some suggests how to choose CNState values or others CNV software which can analysis WGS data?

    Thank you very much !

    Kind regards.

nanshan

[Joe Glessner]

Hi Nanshan,

Most WGS CNV callers provide just DEL or DUP designations because it is difficult to distinguish between the heterozygous state (cn=1 or cn=3) and the homozygous state (cn=0 or cn=4).

I would suggest you can just assume heterozygous CNV calls make up the majority of calls and encode CNState values DEL=>state2,cn=1 and DUP=>state5,cn=3.

If you want to try a more complex approach, you could plot the histogram of RD values you have from CNVkit and look for inflection points that would show appropriate cut-offs for homozygous vs. heterozygous states. However, from my experience the homozygous are so few and differential signal is too weak to separate them.

Here is an awk command that will convert CNVkit output into standard PennCNV format:

awk '{ORS="";print $1":"$2"-"$3" numsnp="$7" length="$4" ";if($6=="DEL"){print"state2,cn=1"}else{print"state5,cn=3"};print" "$5" startsnp="$1"_"$2" endsnp="$1"_"$3"\n"}' CNVkit.txt > CNVkit.rawcnv

Regards,
Joseph Glessner

--
You received this message because you are subscribed to the Google Groups "ParseCNV" group.
To unsubscribe from this group and stop receiving emails from it, send an email to parsecnv+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[nanshan yang]

Thank you very much for your answer!

Best regards

nanshan

在 2018年10月25日星期四 UTC+8上午11:57:28，Joe Glessner写道：