Dear All,
I know pretty well the RNAseq workflow using GenomicAlignment and DEseq2 to analyse RNAseq expression. Now I want to calculate one known fusion gene expression in my two RNAseq samples which are from cancer cells that I know have the fusion gene. So I know this fusion gene sequence, but I have no idea how to count the reads coverage falling on this fusion sequence. Can you help, many thanks in advance !
Stephen
Hi Stephen,
My approach would be to translate the fusion-gene into genomic coordinates en put them in the gtf file which contains the coordinates of all your genes that you use in your analysis. I use featureCount from Subread, followed by edgeR (but I guess the GenomicAlignment and DEseq2 workflow will be similar in principal).
Ben
Thanks. However, I am wondering it this approach works. Since I am thinking that the original fastq reads of that fusion gene were not even mapped to the reference genome as the reference genome does not have the fusion sequence, so the reads of fusion gene likely would not even appear in bam files. If this is the case, does it mean I have to do re-mapping ? And how ?
Stephen
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
The reads from this fusion gene will be mapped to the original genes. If the fusion gene contains the first half of gene A and the latter half of gene B, the reads will be mapped there. Does your sample only contain the fusion gene, or does it also contain wt alleles of both genes?
It of course also depends on what aligner you used, and how it handles for instance split reads and paired end.