FastQ to aligned BAM file Question - Galaxy

[Zapages]

Hi all,

I was wondering, how does bowtie fit in with tophat. What I have done so far is the following through Galaxy-Project.

1) Converted an NGS (RNA-Seq) SRA data to FASTQ file via DRA-SRA or through EBI's SRA to FASTQ converter.

2) Uploaded this FASTQ file onto Galaxy-Project

3) Used FASTQ Groomer

4) Then used FASTQ Groomed file from part 3 with Tophat to get the accepted hits bam file, etc

5) Now based on the workflow that I am following the methods goes back to Groomed FASTQ to do bowtie

but the tophat file and bowtie file never get worked together to make one bam file...

As my ultimate goal is get expression data via cufflinks, how do the two BAM files from bowtie and tophat get incorporated together?

I have been following this: https://main.g2.bx.psu.edu/u/fluidig...naseq-workflow

and

this https://main.g2.bx.psu.edu/u/jeremy/...lysis-exercise

The second one just uses tophat. But shouldn't we index the file via bowtie as well?

If we are just using Tophat, then what purpose does bowtie have then? Because I have read in journals that we should use bowtie and tophat together.

I am confused. If someone could kindly clarify this, I would really appreciate it.

Thank you.

Zapages

[mastal]

FastQ to aligned BAM file Question - Galaxy

I had a look at the Galaxy workflows in your links, and I don't know why they use Bowtie and BWA as well as Tophat.

Bowtie and Tophat are always used together in that, when you run Tophat, it calls Bowtie.

[swbarnes2]

Tophat is pretty much the Bowtie algorithm, but for RNA. It knows that the reads might be spliced, so you can give it a list of exon locations, to help it align reads that cross exons.

[Zapages]

I had a look at the Galaxy workflows in your links, and I don't know why they use Bowtie and BWA as well as Tophat.

Bowtie and Tophat are always used together in that, when you run Tophat, it calls Bowtie.

Tophat is pretty much the Bowtie algorithm, but for RNA. It knows that the reads might be spliced, so you can give it a list of exon locations, to help it align reads that cross exons.

Thank you, that was really helpful. So in other words, Tophat is bascially Bowtie 2 for RNA data sets and it is able to see exon regions for RNA-Seq data.

Am I thinking correctly?

[chadn737]

Tophat uses Bowtie/Bowtie2 for the alignments. It has additional steps to handle splicing, junction discovery, etc. In the Tophat pipeline, Bowtie is just one component so there is no need to do alignments separately with Bowtie. Just stick with Tophat for RNA-seq data and you will be fine.