This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Is There A Rare Variant Burden Test That Corrects For Gene Length?

Hi,

I would like to know if any of the gene rare variant burden tests out there tries to correct for gene length, or number of variants per gene. As far as I know (might be wrong) this doesn´t seem to be implemented in any test. This would lead to always favor significant results for longer genes vs. shorter genes.

Best regards, João

gene length
• 5,106 views • link
updated by Istvan Albert Istvan Albert • written by joaofadista • joaofadista

Thanks for the tip. I will test it out!

• 0 views • link
written by joaofadista • joaofadista

Sorry, but your script does not work if you have less probes per gene.

Let say, substitute your:

n_probes = c(10, 20, 30, 50, 100, 250)

by:

n_probes = c(2, 5, 10, 20, 50, 100)
• 0 views • link
updated by Ram Ram • written by joaofadista • joaofadista

please don't keep adding answers to your question, use the 'comment' link to comment on an answer

• 0 views • link
written by Istvan Albert Istvan Albert

1 answer

I tried this with the CMC test to see if using more probes would change the p-value distribution of randomly generated data (no real signal)

It doesn't look like there's an increase in low p-values after increasing the number of probes in fact it's quite the opposite.

distribution

The code is this

so you can test other burden tests thanks to the AssotesteR package.

• 296 views • link
updated by Ram Ram • written by brentp brentp

can you provide a path to the R package ?! google does not know it ! (probably it is misspelled)

• 0 views • link
written by Quak Quak

Log in to answer this question.

More posts like this