Normalization of gene expression using FPKM values

Question

[mrkumarbiotech]

Dear all,

I'm dealing with RNA seq data and obtained FPKM value using cufflinks for treated and control samples. Do I want to normalize these value to obtain p-value or it is sufficient to find DE genes?. If I want to calculate p value using this FPKM. what would be the best test to go with this? How to get the P value?. So far I have used fold change values and created Venn diagrams to show transcripts between each condition. I need to find DE genes. I am new to this field and So anyone please give some solution for me to move step forward.

Thank you

Answer 1

[Devon Ryan]

Just use cuffdiff with the output rather than doing this yourself. FPKMs should generally be avoided when it comes to performing statistics and cuffdiff does some additional stuff internally to try do statistics in a meaningful way.

[mrkumarbiotech]

Thank u Devon Ryan for your suggestions. Does cuffdiff need replications. I do not have replication values.

[Devon Ryan]

Well, it will run and produce results, but no tool can produce anything reliable without replicates. That's simply impossible.