Question: Which Aligner Should I Use For Mapping Bacterial Rna-Seq Data
gravatar for Kssr
8.2 years ago by
Kssr110 wrote:

The reads are paired end,90 bp. I have tried using bowtie and tophat.Bowtie gave around 85 % of alignment.While tophat gave around 92 % alignment.Although Tophat calls Bowtie, any ideas why the percent alignment is higher in case of Tophat??

As bacteria do not have splice junctions,does running tophat in anyway affect the alignment.How does running tophat with GTF file affect the alignment?

I prefer using tophat since I would like to do differential analysis using tophat output with cufflinks software.Any ideas would be appreciated.

tophat alignment rna bowtie • 7.3k views
ADD COMMENTlink modified 6.4 years ago by Josh Herr5.7k • written 8.2 years ago by Kssr110
gravatar for Ryan Thompson
8.2 years ago by
Ryan Thompson3.4k
TSRI, La Jolla, CA
Ryan Thompson3.4k wrote:

You are probably running Bowtie in paired-end mode, which means that Bowtie requires a pair to either map together or not at all. Tophat maps each read singly because it is looking for splicing events, so it cannot assume that a pair of reads will map near each other because that pair might span an intron.

So the difference in mapping percentage is probably just because Tophat allows reads to map singly. You can do the same thing with Bowtie as well by first running in paired-end mode and using the --un option to collect unaligned pairs, and then mapping the unaligned pairs in single-end mode.

Also, I believe Cufflinks accepts any bam file as input. It is not dependent on Tophat.

ADD COMMENTlink written 8.2 years ago by Ryan Thompson3.4k

Cufflinks does accept any bam file as input, but there is one restriction: there must be XS tags for the spliced alignments.

ADD REPLYlink written 8.2 years ago by Mikael Huss4.7k

Well, bowtie doesn't do spliced alignment, and bacteria don't splice their RNA, so I don't think that should be a problem in this case, either technically or biologically. But good point regardless, for anyone working in eukaryotes.

ADD REPLYlink written 8.2 years ago by Ryan Thompson3.4k
gravatar for Josh Herr
6.4 years ago by
Josh Herr5.7k
University of Nebraska
Josh Herr5.7k wrote:

This is a problem the authors of the Tuxedo suite have touched on, hence, this newish paper: EDGE-pro: Estimated Degree of Gene Expression in Prokaryotic Genomes

ADD COMMENTlink written 6.4 years ago by Josh Herr5.7k
gravatar for Swbarnes2
8.2 years ago by
Swbarnes21.5k wrote:

You could probably just align to a refseq file of transcripts with anything. That would be easy to count from the .bam file.

ADD COMMENTlink written 8.2 years ago by Swbarnes21.5k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1830 users visited in the last hour