We analyzed two publicly available allele-specific binding datasets, one based on the binding of multiple 42 TFs to the diploid personal genome sequence of NA12878, the other based on the binding of CTCF to SNPs discovered through ChIP-seq in 6 different LCLs [30]. We computed the Tv frequency in allele-specific variants and in all variants tested and compared the two frequencies by transforming the assumed binomial distribution to a standard normal and performing a two-tailed Z-test. We pooled allele-specific variants across TFs or across cell lines, making sure to collapse redundant variants. We were ignorant of the overlap of all tested SNPs across cell lines and for simplicity used the mean number of SNPs tested as the null model sample size for that dataset.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.