Interpreting Rpkm Data

Question

[user11235]

I have a bunch of RPKM data obtained experimentally, and I'm looking to interpret this data statistically. This is slightly complicated because a large number does not necessarily mean larger expression since it depends on how many reads we have per sequence. How would you go about comparing RPKM data for different genes while categorization a gene as "highly expressed", "medium expression", and "low expression?"

[Ashutosh Pandey]

Well RPKM is already a normalised score that keeps into account the length of the gene and also total number of mapped reads for that sample. So RPKM values can be compared between two samples. There is no fixed threshold for high, medium or low expressed using RPKM. Different people use different thresholds to classify high or low expressed genes using RPKM. Roughly 1 RPKM denotes a transcript per cell. Read this post: Can RNA Sequencing measure absolute quantities of mRNA transcripts?

Answer 1

[Biomonika (Noolean)]

I would very much recommend you this video: