SNP‐based enrichment analysis for WGCNA modules using an hypergeometric test

Question

[andres.firrincieli]

Hello everyone

I have 13 clusters of co-expressed genes, obtained form an expression dataset of 6800 genes per 27 samples. Among the 6800 genes used for the netowrk analysis, 168 have a single SNPs (compared to the WT strain). I was wondering if an hypergeometric test can applied to find modules enriched with genes having a SNP:

Number of genes with SNP in the turquoise module 17
Total number of genes in the turquoise module 889
Total number of genes in the network 6800
Total number of genes with SNP 168

phyper(17-1,889,6800,168, lower.tail = TRUE, log.p = FALSE) #edited

Does it make any sense?

Thank you

Answer 1

[Kevin Blighe]

I think that it is [almost] correct. Take a look at THIS same question on StackOverflow where it would end up being:

phyper(16, 889, 6800 - 889, 168, lower.tail = TRUE, log.p = FALSE)

Note, also, whether it is 16 or 17? Please re-check your question.

Kevin

[andres.firrincieli]

Thank you Kevin!

I have edited my question