Question: RNAseq normalization for new data
0
gravatar for sosal
9 days ago by
sosal10
South Korea / Seoul / SNU
sosal10 wrote:

Hello biostars!

I want to implement a model that predicts disease from RNA-seq data. Therefore, the training data were normalized by limma voom, and trained by randomForest.

However, new input data has not been normalized in the same way as training data. If I have to normalized the new data with training data, I need to create a new prediction model with new normalized data.

Please, let me know how to normalize for new data in the same way with training data for fixed model prediction.


RNAseq_train # Training data

RNAseq_test # Test data (but can not be combined with RNAseq_train samples)

DesignMatrix <- model.matrix(~0 + Design)
keep=rowSums(cpm(RNAseq_train)>=1)>=keep_n
RNAseq_train_filtered=RNAseq_train[keep,]

DGE=DGEList(RNAseq_train_filtered)
DGE=calcNormFactors(DGE,method =c("TMM"))
RNAseq_train_normalized=voom(DGE,DesignMatrix, plot=F)$E

but how 'RNAseq_test' normalized in the same way with RNAseq_train_normalized? (In the situation that RNAseq_test was new cohort data after training model)

ADD COMMENTlink modified 9 days ago • written 9 days ago by sosal10

Please use the formatting bar (especially the code option) to present your post better. You can use backticks for inline code (`text` becomes text), or select a chunk of text and use the highlighted button to format it as a code block. I've done it for you this time.
code_formatting

ADD REPLYlink written 9 days ago by RamRS21k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 812 users visited in the last hour