This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to calculate correlations across two matrices across samples?

I have methylation count data in the following two matrices.

cov <- matrix(sample(0:7, 100,replace=TRUE), nrow=10, ncol=10)
colnames(cov) <- c("group1_1", "group1_2", "group1_3", "group1_4", "group1_5", "group2_1", "group2_2", "group2_3", "group2_4", "group2_5")
meth <- matrix(sample(7:10, 100,replace=TRUE), nrow=10, ncol=10)
colnames(meth) <- c("group1_1", "group1_2", "group1_3", "group1_4", "group1_5", "group2_1", "group2_2", "group2_3", "group2_4", "group2_5")

First five samples in each matrix belong to group1, second five samples to group2.

I've been asked to look at correlations across the matrices across samples, with the purpose of identifying outliers. How do I perform this test and what should I be looking for?

statistics r sequencing methylation
• 1,828 views • link
updated by Biostar Biostar • written by biohack92 biohack92

1 answer

I think you can cbind the two matrix and do the cluster analysis to check the relationship or distance between the samples in different group and different samples, such kind of Batch effect, outlier and identification can be shown in this way. In addition, you can try some other methods such as PCA, SVD and MDS.

• 0 views • link
written by Shicheng Guo Shicheng Guo

Log in to answer this question.

More posts like this