estimate read counts of a specific region of a transcript

[zixuanxie]

Hi, I'm wondering if there's a way to estimate: for a transcript, the ratio of reads mapped to its 3'end against the reads mapped to its 5'end. Thanks very much.

[Yenaled]

Nope, but here are a few suggestions:

1. In your transcriptome reference, you could split up the transcript(s) associated with the gene into two parts: the 5' part and the 3' part.
2. In kallisto versions (<=0.48.0), you can use --genomebam to create a BAM file that you can load into a genome browser to view where the reads map along a gene.
3. Extract all reads that map to that transcript, and then do a second alignment (using a standard aligner) against that transcript sequence for those reads to see where they map along the transcript.