We are working with noisy (MAD < 0.15, avg. 0.10) low pass WGS data from FFPE blocks where the DNA was sampled from a histologically delineated tumor region.
IchorCNA frequently suggests results with whole genome amplification (dark red). We have done some work on understanding the algorithm for solution selection and have following questions
- Is it correct that the solution is ALWAYS the highest log likelihood after filtering by maxGenomeSubclonal and maxCnaSubclonal?
- Do you have any examples of situations where we should consider changing the default values (maxGenomeSubclonal=0.5, maxCnaSubclonal=0.7) considering the propreties of our data?
We are working with noisy (MAD < 0.15, avg. 0.10) low pass WGS data from FFPE blocks where the DNA was sampled from a histologically delineated tumor region.
IchorCNA frequently suggests results with whole genome amplification (dark red). We have done some work on understanding the algorithm for solution selection and have following questions