search

mcmero / SVclone

A computational method for inferring the cancer cell fraction of tumour structural variation from whole-genome sequencing data.
BSD 3-Clause "New" or "Revised" License
40 stars 10 forks source link

0 spanning read counts #8

Closed trwo closed 4 years ago

trwo commented 4 years ago

Hello,

I'm trying to use SVclone for a list of BRASS calls and the original BAM file. It generates counts and estimates plausible VAFs but the 'spanning' counts column remains empty for all SVs except one. I have visualised all these breakpoints myself in Jbrowse and confirmed they map to other genomic regions using UCSC blat so I'm not sure why it would fail to count them. Do you know of any reason why SVclone might be failing to flag these reads as spanning? It leaves me with a *filtered_svs.tsv file with a single SV rather than 61.

Thanks for your time.

mcmero commented 4 years ago

Please check the read_params.txt file under your output directory. If the fragment length and standard deviation don't look sensible, then this is the most likely cause of your issue and you'll have to manually set them. This can be done by setting these values in the config file under [BamParameters]. Delete the read_params.txt file and try rerunning from the annotate step.

trwo commented 4 years ago

Ah, OK thanks. I will be sure to manually set these.