Trinotate performs very well and in your specific case you should be able to perform a full annotation in a reasonable amount of time. First, you need to run TransDecoder, which is rather quick, then you can proceed with BLAST, which is the most time-demanding step of the annotation. You can choose between BLASTp with the TransDecoder predictions (quicker way) or BLASTx with the assembled contigs (slower but somewhat more sensitive, especially for partial contigs and transcripts encoding short proteins). You may also want to use UniprotKB/SwissProt instead of nr as a reference database for BLAST to speed-up your analysis (but this choice depends on the organism you are workign with, if it is a vertebrate I'd say running the anotation step on the nr database is a waste of time and you can safely use UniprotKB instead).

The PFAM annoation with HMMER is rapid and GO terms are inferred from BLAST hits and Pfam domains.

The only real problem with Trinotate is that the input multi fasta file has to be generated with Trinity, so if your de novo assembly has been performed with another assembler you will need to rename all the contigs according to the Trinity contig nomenclature scheme (if this is your case I will need to give you an additional heads-up).

Blast2GO basically does the same, but in my experience it is much slower (unless you run your analyses locally, but I think some features are limited to the Pro version).

Thank, Marco. My organism is unicellular organism, in fact dinoflagellates, which has huge genomes.

If you nave noto used trinity follow the suggestion of Marco. Should not take a long time. It is simple and could be a good solution.

We have recently published our trnascriptome annotation pipeline to detect coding as well as long non-coding RNAs in assembled transcriptome.

http://www.ncbi.nlm.nih.gov/pubmed/25701574

Available at: https://github.com/frankMusacchia/Annocript

Once configured the pipeline runs locally and is tailor made for large transcriptome datasets.

Hi everyone, I' am trying to use Trinotate with a non-Trinity assembly but I couldn't make. Anyone has a script to help me with fasta header formatting. I am not sure how to change the fasta header to make them compatible with Trianotate. Any help would be really appreciate. Thanks

Hi everybody

I am a verdant young scientist. I came across this post while trying to find a solution to a similar problem I am having, when trying to run Trinotate on a non-trinity assembly. I am interested in knowing whether there is a way to run Trinotate on a non-trinity assembly. I would be so appreciated in advance if any one could show me a right way or how I can get around it. I am keeping my eyes open to see your comments. Thanks

Hi Abozaar, simply change the name of the contigs in your assembly so they will match those expected by Trinity. In the latest version, thy look like:

TRINITY_DNX_c0_g1_i1, where X is the number of your contig

So, let's say the original names of your contigs are:

contig1

contig2

contig3

etc.

They should be renamed as

TRINITY_DN1_c0_g1_i1

TRINITY_DN2_c0_g1_i1

TRINITY_DN3_c0_g1_i1

As long as your assembly does not contain alternatively spliced isoforms, don't bother about the "c0_g1_i1" part and leave it as it is.

This should work fine. No need for scripting, you can simply do that in a text editor by using regular expressions.

Hi Marco

Thanks for your speedy response. I will try and keep you posted how it goes 

Thanks Marco. It works