Identifying full length transcripts in a transcriptome

Question

[EVR]

Hi,

Is there any way to find full length and in-complete transcripts that has been assembled from a de novo transcriptome?

Any guidance would be appreciated.

Thanks in advance

[trausch]

The CEGMA benchmark (PMID: 17332020) evaluates how many core eukaryotic genes have been assembled partially or full length. A similar tool seems to be BUSCO. I am not sure if this can be generalized beyond core single-copy genes.

[iraun]

Do you mean a way to calculate de length of sequences in a multifasta file?

[Devon Ryan]

I suspect Tom instead wants a way to see if the assembled contigs are complete/full or not.

[EVR]

Like Devon Ryan said, I want to find whether assembled contigs in a the transcriptome is complete or missing some parts? Can we BLAST the query transcript and check and how good is the coverage for the best target hit. And later add the missing region from the target best hit which is not present in query? is my approach is right? Kindly guide me

Answer 1

[JC]

Run Triannotate https://trinotate.github.io/ and check how many are complete or not (the fasta header has a tag for that)

[EVR]

You mean to say "Transdecoder" cos trinotate annotates whiel transdecoder predicts the protein coding transcripts based on ORF