Does Bowtie2 Map Multiple Reads To The Same Gene?

Question

[derekc10741]

Hi all, I'm new to bioinformatics and I've a question about bowtie2. I know that bowtie2 reports read that map to the reference genome, but what if there are multiple reads that map to the same exact same gene? Does bowtie2 support that? I know it looks for the best alignment, and its decision of which alignment to report is random, but if there are 5 reads which all align equally well to a gene in a reference genome, does it's output only have 1 of those reads, or does it have all 5? Thanks

[KCC]

Perhaps you are already aware: however, I thought I'd mention that the aligner (bowtie 2) is not aware of genes.

[Ryan Thompson]

I think you have it backwards. Did you mean to ask what happens if the same read aligns to multiple locations in the genome?

Answer 1

[Simon Cockell]

The best alignment for each read is reported independently. You can have many reads mapping to exactly the same genomic coordinates (and indeed often will, especially with particular experiment types) and they will all be reported in the SAM file.

[deepti1rao]

And then how do you go about deciding which read is the best match if mapQ is also the same for all?

Answer 2

[Rahul Sharma]

Hi,

You can use the read depth viewers(IGV, tablet or IGB) and see the coverage depth of your gene. You need to import the reference genome file, Its gff3 file and SAM file generated by bowtie2. Or one can also try samtools mileup(http://samtools.sourceforge.net/mpileup.shtml) it will generate the read coverage output in tabular form for all genome positions, later unix awk/grep can be used for further analysis.

Regards, Rahul