Entering edit mode
4.4 years ago
goodez ▴ 630
I have aligned GRO-seq (think ChIP-seq) reads using STAR in splice-aware mode. Will my reads still align correctly if they cross an exon-intron boundary. I would think they should. This may seem like a strange question and perhaps I should have used bowtie2 or something splice-unaware. But I know STAR is pretty fast and this is a huge alignment.
There was a discussion on how to use STAR on transcriptome annotation here. With these settings, you shouldn't get any spliced reads (or those having indels).
You can have a look with IGV how the reads are aligned or use RSeQC to check the reported junctions.
What do you mean I shouldn't get spliced reads? I already have the reads, which come from DNA sequence so nothing is spliced. I really need to know how STAR handles aligning unspliced DNA sequence across an exon junction. Since, of course, STAR is expecting reads to be spliced, but will it still correctly align across the junctions?
I'd check if you have spliced reads or not.
In my experience, STAR aligns the reads correctly; even over exon-intron boundaries. But I'd be rather on the safe side.
Thank you. I'm assuming you mean check if I have spliced alignments? Because my reads are most definitely not spliced. I actually started a bowtie2 alignment. So I guess I can compare the STAR and bowtie2 results and report back later.
Could please share your experience of using star for GRO-seq data alignment and how did it compare with Bowtie2? Thanks
See this thread for caveats and suggestions:
align genomic DNA using STAR