Assembling a Transcriptome

If you are studying an organism that does not have a well-documented transcriptome, you can still study gene expression by assembling your RNASeq reads into a set of transcripts or a transcriptome. Even if you work with an organism that has a sequenced genome and some RNASeq data available, it may be important to your studies of specific tissues or of specific conditions may require generating your own, new set of transcripts. There are different approaches to doing this, but all will start with RNASeq read files that have been cleaned up as, for example, described here: Cleaning up your reads with the HTProcess Pipeline. 

Guides to Transcriptome Assembly

1. A tutorial on de novo transcriptome assembly using Soapdenovo-trans: http://www.iplantcollaborative.org/learning-center/discovery-environment/de-005-transcriptome-assembly-de-novo.
2. A webinar and related materials on de novo transcriptome assembly: iPlant Webinar- Transcript Assembly Approaches
3. A workflow for de novo transcriptome assembly: Denovo Transcriptome Assembly Workflow
4. Another consideration for the user: how big do you think your transcriptome is and how many fastq files do you have? If you have a relatively small job, e.g. 6 moderate (e.g. 4 GB fastq files or less), consider using Soapdenovo-trans-small instead of Soapdenovo-trans. They are the same app except the small version runs on the fast new Lonestar 5 server, and is limited to 64 GB ram. Soapdenovo-trans seems better than Trinity at managing memory usage, so you have a good chance of getting your job done quickly and you will be saving the load on our largemem nodes on the Stampede server.
5. Reference Guided assembly – for those who have a genome sequence to work with: Reference Guided Transcriptome Assembly.

Related articles