Regions of excessive sequencing depth for hg38

[zhujack]

Hi,

I tried to generate Regions of excessive sequencing depth for hg38, but the UCSC table browser does not have the Top 0.01 frequency Hi Seq Depth dataset for hg38. It was suggested on the WiKi page to use the detectExtremeDepth utility from cgpBigWig to generate the file. I wonder how exactly it should be generated with detectExtremeDepth utility. I have tried the following:

bam2bw -i $tumor -o tumor.bw
detectExtremeDepth -b tumor.bw -o depth_out

Is this the right direction? Could you give me more detailed instructions on this? Thanks.

Jack

[AndyMenzies]

Hi Jack

You are going in the right direction. I have updated the BRASS wiki page with a link to the right page in PCAP-core.

https://github.com/cancerit/PCAP-core/wiki/Hi-seq-depth-regions

This will take you through the process.

• Its better to combine the results of multiple samples to better identify high depth regions.
• You need to make sure the samples used for this are representative of the samples you will be processing. Ie, same sequencing platform, same read mapping algorithm.
• It also tends to be better if you use normal samples to build the filter, if a tumour sample has an extreme copy number change it might get included.

Hope this helps

Andy