Using Galaxy/Bowtie/Cufflinks/Cuffdiff for Illumina RNA seq on bacterial transcriptome (E.coli).

[Mathia Arens]

Hi all,

I am trying to analyse my RNA-seq data (mRNA from E.coli) from the Illumina Miseq. I would really like to use Galaxy to analyse my data, because I have no experience with programming. I have seen that Galaxy is used more for eukaryotic transcriptomes, however Galaxy states that it is possible to use it for bacterial transcriptomes as well.

My workflow looks like this: Illumina .Fastq file --> fastq groomer to get fastqsanger file --> Map with Bowtie --> SAM to BAM --> Cufflinks --> Cuffdiff --> visualization

The steps up until SAM to BAM seem to work fine, but I get stuck on cufflinks.

Does anybody know for a bacterial transcriptome what to fill in for the parameters: max. Intron length, min isoform fraction and pre mRNA fraction?

These parameters all seem to be applicable for eukaryotic RNA, but not so much for prokaryotic.

I hope someone can help me with this, would be very much appreciated.

Kind Regards,

Mathia

[Steven Salzberg]

We really never planned for people to use Cufflinks for bacteria, but we did publish a separate system for this purpose last year:

http://ccb.jhu.edu/software/EDGE-pro/index.shtml

EDGE-pro will assign reads to genes and produce FPKM/RPKM values for bacterial genomes.  It doesn't have much of the functionality of Cufflinks, though - in particular it does not do differential expression.  However, you can take its output and use that as input to Cuffdiff (or other tools) if you need that.