search

bartongroup / yanocomp

Yet another nanopore modification comparison tool
MIT License
11 stars 1 forks source link

How can I get p values for all the positions? #9

Closed bhargava-morampalli closed 2 years ago

bhargava-morampalli commented 2 years ago

The default result is for most significant position in the RNA but is there an option which I can use to get the p-values for all the positions?

mparker2 commented 2 years ago

Hi @bhargava-morampalli, yes you can use the option --fdr-threshold to relax the filtering. However, positions are pre-filtered using a KS test to reduce the number of positions which have to be tested with the GMM method. The KS test uses the same --fdr-threshold parameter as well as the --min-ks parameter to filter out sites that are not worth testing. So you would need to relax both e.g. by setting --fdr-threshold to 1 and --min-ks to 0, to get results for all positions in the transcriptome. BUT this would increase the number of GMM models that need to be fit greatly and so will potentially increase the run time by quite a lot. I also have no idea how it will affect the false discovery rate for the GMM test.