I have 75 samples' data downloaded from TCGA. While running some of the steps in the pipelines like ChAMP, Minfi, etc., I get maximum memory allocation reached error. I have 4 GB RAM and i3 processor. Please suggest me possible alternatives or workarounds? Thanks much.
4 gigs of RAM isn't exactly a lot these days, particularly when you're talking about manipulating datasets of the size you're using. Try running this on a larger server.
I definitely told you that it is the problem of memory. Methylation 450K array is very large data actually. It is very easy. The minimum memory should be > 450K*75 = 33.75G, Usually, I will do DMR analysis in memory=512G PCs. Such 512G memory can keep you do PCA, SVD, Cluster analysis, Classification and some other traditional data mining operation. If you have problems from memory, you can separate the data by the chromosomes and deal with them chromosome by chromosome.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Oh all right! I just thought maybe I could make-do by tweaking the code a little bit, which might increase the computational time. Is it possible? Could I break the process into smaller parts and solve this issue?
That I don't know, you'd need to look into exactly how the individual functions work. The biggest issue will be in any normalization step, where having the whole thing in memory is typically needed.