Possible to map reads against multiple separate references at once?

I have a project wherein several known templates were used to generate an Illumina library. Once the pool is sequenced, I need to count the abundance of each read that maps to (originated from) its source template.
Some considered options:
(1) perform OTU analysis on the groups, then ID each OTU and count;
(2) link each source templates together and perform a ‘standard’ (BWA/Bowtie) map and count read depth at each position;
(3) map the reads independently to each reference and count (laborious, serial operation)
(4) map the reads to a reference that contains the template sequences as separate entities, then get a count per reference.

Option 4 would be the most straightforward, but I don’t think there’s a way in index separate references, so I am leaning to option 2.

Considering the tools on Galaxy.us - can someone provide feedback on the proporosed approaches, or provide an alternative?

Hi @sdmoore

Any of these could be done. We have protocols for OTU analysis (1). Repeating a multi-step workflow across samples is somewhat trivial (3). And creating your own reference sequence (2, 4) is definitely possible.

I’m going to list some resources to start with that can maybe help frame any followup questions you may have.

For an overview, please start here to learn how the Galaxy platform used. In particular, see the section Applications then maybe the Community special interest groups.

Then you can zoom into tutorials that might do something similar to use as a guide, and you can find workflow templates here as well. Pathways could help, or review by domain or your read type.

If you are completely new to Galaxy, this is a nice simple introduction that includes a workflow so you can see how those can help with repeating or tuning the same series of manipulations on data.

And, you can work at any of the UseGalaxy servers, or any of the others. These host both common and distinct resources and moving data around between them is easy. What you plan to do is part of deciding where.

Followup questions are welcome. What kind of read data do you have? What are those templates? Do you have a reference publication already that does something similar that you want to do or that you used to guide the sequencing strategy?

Let’s start there, thanks! :slight_smile:

Perfect, much appreciated.

1 Like