choose the "best"mutation from a list

Question

[fifivater]

I have a list of mutations (around 400) from COMIC database. I have given a score to each mutation from different tools e.g by position, conservation, blosum, patogenecity...for each mutation position. My goal is to find a "easy" mathematical model or a model to choose from this list the "best" mutation (around 10 mutants) to take to cell culture lab. Could you help what model I can use? I was thinking something like nX+nY+nZ. X, Y, Z are the score from different tools I got and n would be 1/3

Answer 1

[WouterDeCoster]

What does it mean for a mutation to be the best? Do you mean "the most likely pathogenic"?

[fifivater]

Yes, I mean the most pathogenic mutant. So I have calculated score for each mutants (~300) by different tools e.g. Blosum, PON-P2, ConSurf. Also i have given a score by position. The Scores from each tool has been normalized betwen 0-1. So my plan is to find a simple mathematical model to choose from this list ~10-20 pathogenic mutant to take to the next level. I need somehow take consider scores from each tool, and find what final score it will give me.

[WouterDeCoster]

I would suggest having a look at tools like CADD, fathmm or DANN

[fifivater]

Thank you very much for the hints, I will get familiar myself with those tools Thanks :)

[Kevin Blighe]

More tools listed here: A: How to choose good tools for identifying functional SNPs?