Open EKocakavuk opened 3 months ago
Hi Emre,
Thanks for your feedback. I wrote this piece of code several years ago, when I experimented with ASCAT for WES samples, and I must admit that I have forgot about the details.
Back then, I used it on many samples, and it never failed. Maybe I did some pre-filtering of sites before running gatk2ascat
, but I think that I just followed a tutorial on the GATK webpage.
I'm sorry that I can't help here, but I suggest that you go ahead with your changes and see if the final output makes sense. Personally, I was never really happy with ASCAT, and these days I'm working almost exclusively with WGS, where I've found Sequenza to be a much better alternative.
Best, Michael
Hi @micknudsen,
thanks for providing your code for the transformation of GATK outputs as inputs for ASCAT. I am trying to adapt this to our current pipeline in which we analyze WXS/WGS data with trios (normal sample + two timely separated tumor samples).
I was encountering some errors when running your script, specifically related to
Since the allelic counts files for tumor and normal include positions where alt_count and ref_count are zero, it gave a non-dividable by zero error, which I attempted to fix with:
Another point is related to the following part:
-> Since not all positions of the allelic count files are included in the denoisedCR file, which throws the error
UncoveredPositionError
I attempted to fix this is as following:
Please let me know your comments. I am not sure what specific down-stream effects these modifications might have and your feedback is appreciated.
Best, Emre