I have two samples, one is drug-treated and the other is not. I ran MiSeq on my samples and generated FPKM values. Some other author had done similar studies earlier and I want to compare my FPKM values with the author's RPKM values. Are these two types of values the same? Can I compare both values?
With single-end reads, FPKM and RPKM are synonymous. With paired-end reads RPKM = 2 x FPKM.
Maybe the safest thing is to retrieve that authors original reads from the repository, or from the author, and realign them to the genome using the same pipeline and settings you used for your data. If you use different alignment parameters or algorithms, then the two experiments would not necessarily be directly comparable - if the alignment methods are significantly different you might need to account for alignment efficiency, which might be sequence-dependent in a different way in each method (non-linear).
You could probably safely compare log fold change in your experiment with log fold change in the other experiment to search for correlation though.
With single-end reads, FPKM and RPKM are synonymous. With paired-end reads RPKM = 2 x FPKM.
Maybe the safest thing is to retrieve that authors original reads from the repository, or from the author, and realign them to the genome using the same pipeline and settings you used for your data. If you use different alignment parameters or algorithms, then the two experiments would not necessarily be directly comparable - if the alignment methods are significantly different you might need to account for alignment efficiency, which might be sequence-dependent in a different way in each method (non-linear).
You could probably safely compare log fold change in your experiment with log fold change in the other experiment to search for correlation though.
FPKM and RPKM are both normalisations for WITHIN sample comparisons, not BETWEEN samples. You will need to do some additional normalisation if you want to make comparisons and/or calculate differential expression.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology