Reads coverage along genes in RNA-seq

[a.kmg]

Hello everyone,

In RNA-seq, it seems that the coverage of reads along the gene is not uniform.

Indeed, it is possible that reads covering is more high in 5', 3', ...

There is a way of knowing the extent of this bias? For example, the percentage of length covered for each gene or a percentage of variability ? Someone already worked on this subject ?

I use the "geneBody_coverage.py" script of RSeQC tool, but it represents an average of the cover along genes, on all genes. This is not accurate enough.

[syfo]

A tricky point is to define the precise criteria you want to consider (what do you mean exactly by "the percentage of length covered for each gene" or "a percentage of variability"?). A measure that would be both intuitive and easy to compute could be the ratio between the read coverage in the first vs. the last exon for each "gene" (or more exactly for each annotated transcript), but you might think of more sophisticated metrics. Maybe this could help: "Comprehensive comparative analysis of strand-specific RNA sequencing methods"