How To Perform Automated Structural Annotation In De Novo Transcriptomes

Question

[pld]

I am interested in detecting structures in a de novo transcriptome I am in the process of annotating. I am interested in seeing if there are any tools that I could use to rapidly identify structural motifs in my transcripts, especially those that do not have orthologs in other species. I am expecting a large amount of data so I was wondering if anyone has any experience doing this sort of annotation in a programmatic way. What would be a reasonable place to begin?

Answer 1

[francescomusacchia]

You may want to use Annocript. It was just published on Bioinformatics and permits annotation of any type of transcriptome (both nucleotides and peptides). It is a pipeline that runs BLAST programs and gives the best HIT against SwissProt and TrEMBL or UniRef, CDD domains, rFam. Annotation also provide GO terms association with proteins and/or domains, Enzyme and Pathwyas. Moreover it also presents which transcripts could be putative lncRNAs. It gives a lot of outputs: a complete table with the annotation, plots and statistics, blast output in GFF3 format, fasta file with the putative coding and long non coding sequences.

The program is straightforward to use and open source. Here is the paper, and here is the software.