I have some question about the CopywriteR package:

1) I have observed that CopywriteR did not detect any copy number variations (CNVs) in normal samples. Is it possible to use CopywriteR for CNV detection?

2) What’s the large gap inside chromosome?(e.g. chr1,chr9) is this regions “blacklist”?

3) Is there any cut-off value for judging whether CNV exist or not?

4) Can I extract candidate genes or sites of amplification or deletion using your algorithm? 

Lim

Regarding your question about analysing the normal sample: I would not expect any amplifications (copy number alterations), but only (germline) copy number variations. The CopywriteR tool is written for detection of copy number alterations (for instance in tumor material), and it filters out CNV regions using the blacklist filter. If you are interested in CNVs, you can quite easily create a blacklist file (the blacklist.Rdata file in your hgxx_100kb folder) with only one line (for instance chromosome 1 start 1 end 1). This will effectively leave any CNV region and would allow you to judge CNVs.

With regards to the gaps in chr1, chr9: these are repetitive regions around the centromere to which mapping is not possible, and hence copy number detection is not possible either. Similarly, the rDNA regions on the acrocentric chromosomes 13,14,15,21 and 22 are lacking too due to their repetitive nature.

Finally, in copy number variation analysis usually multiple steps are taken towards an end result. First, reads are pre-processed, mapped, and copy number information is retrieved, then copy number info is segmented (in our case using DNAcopy), and finally one can perform what is known as ‘calling’. Here, an absolute estimate is made of what is gained and lost (and sometimes the absolute copy number is estimated too). I wrote some comments on it on this: A: Calling amplifications and deletions from CopywriteR output. I hope you find useful info to get started there.

Thomas