search

friend1ws / SAVNet

detection of mutations causing splicing change
GNU General Public License v3.0
13 stars 4 forks source link

How to deal with multiple samples per patient #2

Open suhrig opened 5 years ago

suhrig commented 5 years ago

Dear Yuichi,

I am working on a cohort of patients with several samples per patient. The samples were obtained from related specimens (metastases/relapses), so a notable overlap in the genomic variants is expected. What is your recommended strategy to deal with such cases when looking for aberrant splicing using SAVNet? I suspect if I treat samples from the same patient as independent, SAVNet might be tempted to overestimate the significance of genomic variants in these samples, because many variants are recurrent.

Thanks, Sebastian

friend1ws commented 5 years ago

Thank you very much for your interest in our software. I basically agree with your concern that each genomic data is not independent and some overestimation problem may occur. But at the same time, just by my experience, I think SAVNet may produce reasonable result even that case. Do you have RNA-seq for each genomic data? In that case, I think you can perform SAVNet straightforward first. So please try first.