Dr. Smyth, Thank you for answering my question a while ago. Now I just have a quick follow-up question for neqc() that needs your confirmation. When combining four chips of microarray experiments (four data sets) into one data set, I agree that quantile normalization of neqc() is good enough as I found the boxplots of the expression values for every sample after normalization were in almost the same distribution. But I saw someone use lumi package to analyze similar data with batch effect removal using Combat. Therefore, I am wondering if I need to consider batch effect removal besides of the neqc() step when using limma?? The 4 chips of microarray experiments were done using different samples on different days. Thanks a lot in advance! Best, Xiayu -----Original Message----- From: Gordon K Smyth [mailto:smyth@wehi.EDU.AU] Sent: Friday, June 28, 2013 2:37 AM To: Rao,Xiayu Cc: Bioconductor mailing list Subject: read.ilmn() and variation between chips Dear Xiayu, Yes, it is good enough. neqc() has done between-array normalization already, and there is no need for within-array normalization for Illumina BeadChips. Please look at the help page ?neqc The read stages that you describe sound complicated. read.ilmn() reads the files as output by Genome Studio at our core facility without any need for intermediate processing. Best wishes Gordon -------------------- original message -------------------- [BioC] read.ilmn() and variation between chips Rao,Xiayu XRao at mdanderson.org Wed Jun 26 20:08:09 CEST 2013 Hello, I have a question about background correction and normalization. Please help me out! Thank you for your time! I have four chips of microarray experiments, and therefore four data sets. I combined them together by merging on ProbeID and read in them as one file using read.ilmn(), and I combined all the control probe files into one and read it in. I just followed the limma user guide and use neqc() for background correction and normalization. Is it good enough? Do I need to consider within array and between array normalization? Thanks, Xiayu ______________________________________________________________________ The information in this email is confidential and intend...{{dropped:6}}