What protocols / options do you use for your Drosophila RNA-seq analysis?

[scottdaniel]

After reading this about Tophat:

From the Tophat manual

Please Note TopHat has a number of parameters and options, and their default values are tuned for processing mammalian RNA-Seq reads.
If you would like to use TopHat for another class of organism, we recommend setting some of the parameters with more strict, conservative values than their defaults.
Usually, setting the maximum intron size to 4 or 5 Kb is sufficient to discover most junctions while keeping the number of false positives low.

And perusing this article:
RNA-Seq: The Drosophila melanogaster transcriptome by paired-end RNA sequencing.

I'm all confused about what best to use for Drosophila RNA-seq analysis. Do any of you use some standard options in your Tophat workflow? Do any of you use BLAT / SOAP / etc. because it is superior to Tophat when it comes to Drosophila?

Thanks.

[Hugo A F Santos]

After reading this about Tophat:

From the Tophat manual



And perusing this article:
RNA-Seq: The Drosophila melanogaster transcriptome by paired-end RNA sequencing.

I'm all confused about what best to use for Drosophila RNA-seq analysis. Do any of you use some standard options in your Tophat workflow? Do any of you use BLAT / SOAP / etc. because it is superior to Tophat when it comes to Drosophila?

Thanks.

-----------BUMP-----------

Dear all, sorry for bumping such an old post... but then again, for me it doesn't feel outdated at all; how can one infer about the most correct properties to use in the TopHat2/Bowtie2 algorithm in order to adapt it to non-mammalian genomes?

Thanks in advance for your feedback.

Best,
Hugo

[scottdaniel]

-----------BUMP-----------

Dear all, sorry for bumping such an old post... but then again, for me it doesn't feel outdated at all; how can one infer about the most correct properties to use in the TopHat2/Bowtie2 algorithm in order to adapt it to non-mammalian genomes?

Thanks in advance for your feedback.

Best,
Hugo

Hey, thanks for bumping such an old post. One thing that I've found out is that flybase used Tophat themselves for the original RNA-seq data on development stages and tissue but in there newer papers they've begun using STAR aligner. I myself am still struggling with Tophat but haven't really modified any parameters to make it more "conservative" as they say in the manual. I've tried STAR aligner and it works quite easily. It's fast but the drawback is that I've gotten reduced alignment rates (about 70-80% with tophat and 30-50% with STAR aligner).

What have you tried so far with your project?

[Brian Bushnell]

If you are having trouble with low alignment rates on RNA-seq data, I recommend trying BBMap; it has much higher sensitivity than Tophat2 or STAR. The only parameter you need to change from the defaults to do RNA-seq on animals is to add "maxindel=100000"

[cascoamarillo]

I would also modify "-I/--max-intron-length", which has a defaul value of 500000. It can be useful for mammals; but try to see the intron size of you species (average, max, min) and use it for your tophat analisis.

[scottdaniel]

I would also modify "-I/--max-intron-length", which has a defaul value of 500000. It can be useful for mammals; but try to see the intron size of you species (average, max, min) and use it for your tophat analisis.

Good idea. I think I had tried this with STAR aligner. Although it didn't improve much there it might improve things with tophat.

[alexdobin]

Hey, thanks for bumping such an old post. One thing that I've found out is that flybase used Tophat themselves for the original RNA-seq data on development stages and tissue but in there newer papers they've begun using STAR aligner. I myself am still struggling with Tophat but haven't really modified any parameters to make it more "conservative" as they say in the manual. I've tried STAR aligner and it works quite easily. It's fast but the drawback is that I've gotten reduced alignment rates (about 70-80% with tophat and 30-50% with STAR aligner).

What have you tried so far with your project?

Hi Scott,

in my experience STAR usually maps more reads than TopHat. Could you post the Log.final.out file as well as a summary from TopHat (or samtools flagstat on the TopHat BAM)? One possibility is that many reads cannot be mapped as concordant pairs - those are output by TopHat but not by STAR.

Cheers
Alex