Question: Can loess.normalization be used for normalizing RNASeq performed with two ERCC mixes?
gravatar for manjovial
2.1 years ago by
manjovial0 wrote:


I am working on RNAseq data where my colleague has used two ERCC mixes , Mix 1 for tests and Mix 2 for WT. So, I have 92 ERCC transcripts with different expected fold change. I am not sure how to provide the information for two mixes when I normalize the data using loess.normalization. I know I should provide a subset of rows (ERCC transcript) and perform normalization to the whole matrix in R as shown below.

result <- format(loess.normalize(mat,mat[15337:15427,1:4], epsilon = 10^-2, maxit = 1, =TRUE, verbose = TRUE, span = 2/3, family.loess ="symmetric"),digits=2,nsmall=2)

where, mat is the matrix of FPKM values for WT and three different tests, mat[15337:15427,1:4] are the rows corresponding to ERCC FPKM.

I have two questions: 1) Does loess.normalize normalizes other genes of the matrix column wise corresponding to the FPKM values of ERCC for that column or it also takes care of fold change 2)How to provide fold change information, if it doesn't consider fold change

Thanks for any help.

rna-seq normalization ercc • 1.1k views
ADD COMMENTlink modified 13 months ago by Kevin Blighe35k • written 2.1 years ago by manjovial0

Not sure about the answer, but maybe you can find a hint at ?

ADD REPLYlink written 13 months ago by Charles Plessy2.6k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 824 users visited in the last hour