Possibility to retrieve p-values?

[Joannavonberg]

Dear developers,

Thank you for your help in installing Intogen. I've been able to run it without problems! I do have some questions. I am wondering whether it is possible to retrieve the p-values for each gene, before they were converted to q-values. Furthermore, I know that some of the methods do their own filtering so it might not be possible, but ideally I would like to have a p-value for each gene (so also for those that were not significantly enriched for mutations). Is this already possible for the user, or is this something that could be added as an option to Intogen?

Best,
Joanna

[FedericaBrando]

Hi @Joannavonberg ,

Thanks for reaching out. I'm glad you were able to run IntOGen successfully! If you want to check specific output of single methods you can inspect more deeply intogen output.

In fact, in your output folder, you can find a folder named steps that contains all the outputs from each of the step of the pipeline. Here you could inspect the results of the methods and their own pvalue computed per gene (in most cases). Be aware that the p-value for the single method may not reflect the final output of IntOGen because tof the combination step.

The combination takes into consideration the output of all the methods for that specific cohort and detects if a gene is a driver combining this knowledge. I advise you to read the documentation to get a full insight of what is going on there [combination]

[FedericaBrando]

Hi @Joannavonberg , Have you found what you were looking for? Please if you need anything else, let us know - Otherwise I will proceed to close the issue as completed! :)

[Joannavonberg]

Hi @FedericaBrando, I'm so sorry for my late reply! I completely missed your initial response in my mailbox, so I'm only seeing this just now.

I know about the steps folder, and am familiar with how Intogen works. But thank you for answering so elaborately!
I was asking about the final results, after the combination step. Intogen only reports the q-values in the final results, while we would be interested in also obtaining the p-values (before the FDR-correction that leads to q-values). The second part of my question pertains to the fact that Intogen, in the unfiltered_drivers file, does not give a p-value for each gene, so there is still some filtering going on. Now I know that some methods can not calculate a p-value for each gene (for instance OncodriveFML, which needs mutations with an impact score to be present in a gene, otherwise it cannot do anything). So I don't know if that is the reason that genes are missing in unfiltered_drivers, or if there is additional filtering happening. I hope I made myself a bit more clear ☺