detecting overlapping CNV from 2 algorithms

Question

[Myosotis1979]

Dear all I used PennCNV and R-GADA to detect CNVs on a set of individuals. In order to detect overlapping CNVs from both algorithm in the same individual (thus confirming that is a "confident" segment), I used the intersect tool of bedtools. I found only one CNV that was detected by both algorithms in the same individual. So I don't think that is a reliable result. I was wondering if combing all the CNVs from the 2 algorithm is a good idea or should I use only the output of one of them. I would greatly appreciate your advices. Thanks

[h.mon]

How many CNVs each algorithm detected? If only a few CNVs have been predicted by each algorithm, finding few overlaps would be expected. What kind of data you have? Do you have good coverage?

What was the exact command-line used to check the intersection between predicted CNVs?

[Myosotis1979]

Thanks for replying. PennCNV detected more than 3000 and R-GADA more than 1100 on SNP genotype data of Affy 6.0 array. I have applied stringent criteria : CNV were kept if length is ranging from 1kb to 3Mb and having more than 10 SNP . CNV overlapping with centromeric, telomeric and immunologic regions were discarded. Regarding the bedtools command , I tried this one (without -f option): bedtools intersect -a penncnv.bed -b gada.bed -wa -wb Thanks :)