RNA-seq and PET-seq

[bogdan]

Hello everyone.

The question may have been answered already : I am looking for a list of packages I could use for Illumina paired - end sequencing analysis or/and for mRNA sequencing analysis.
Any suggestions would be much appreciated. Thanks,

thanks very much,

bogdan

[Colinmt]

We've been using the CLC bio genomics workbench software package. Downside is you have to pay for it. Upside - its very, very comprehensive and should cover all of your needs.

[bogdan]

free software

Thanks Colin. I would prefer to use some free software though, if it is available. Thanks !

[Melissa]

Most software that can handle illumina reads can support PE reads. I'll let other more experienced members to give you the list. Don't worry. There's a long list of freeware. It depends on what kind of mRNA-seq analysis (de novo assembly, gene expression, microRNA etc?).

I haven't heard about any that is good with RNA-seq de novo assembly. Any idea?

Melissa

[bogdan]

Hi Melissa,

we are doing mRNA-seq. There are some packages as ERANGE and RSAT :
did these work very well? thanks,

bogdan

[Cole Trapnell]

You might try my TopHat tool (http://tophat.cbcb.umd.edu), which now supports both paired ends and longer read lengths. It's still in "beta", but you might find it helpful. It will align RNA-Seq reads across splice junctions without needing a reference annotation (but if you have one, you can use it).

[bogdan]

TopHat

Hi Cole,

yes, I did install the package you did make. it works great on the test data, and will apply to our data soon. A question though : are there any options/parameters I shall use in order to
output the quantitative expression instead of the splice junctions ? thanks a lot,

bogdan

[Cole Trapnell]

You'll need a GFF file for the RPKM calculation, but as far as parameters go, the defaults should be fine if you are working with a mammalian genome. If you're not, I'd recommend dropping down the maximum ab initio intron length (the -I param) to a few Kb for your first run, just to get a conservative idea of how TopHat performs.

If you don't want TopHat to look for junctions ab initio at all, you can use --no-novel-juncs to put TopHat in a "validate only" mode, where it will just try to align your RNA-Seq reads across the junctions in your GFF file (or your raw junctions file if you use the -j option).

[bogdan]

Thanks, Cole. Will proceed accordingly. If I have any other questions, perhaps I can contact you directly ? And congrats for the work ! The packages and the algorithms are designed in a very professional way !

[Cole Trapnell]

Yes, please feel free to contact me directly (cole AT cs DOT umd DOT edu). Good luck!