Time to align full-length transcripts against a reference

Question

[pablo]

Hello,

I will have to align PacBio full-length transcripts against a reference.

I'm gonna use the isoseq3 pipeline in order to map my full-length transcripts against my reference, is it the good workflow?

My other question is : do you have an idea of the necessary time to assemble a 500Mb diploide genome with the smrtlink package? Or should I use another package like canu or hifiasm?

Bests

Answer 1

[colindaven]

This is strongly dependent on your infrastructure. How big is your cluster / or single server ? Why not just try it ? I haven't used isoseq3 but am a big fan of gmap with the GFF3 option for transcript to reference mapping. Evaluation by eye in a web browser is critical.

Generally though, assembly will take a lot longer than mapping. Canu can take weeks for 3 GB haploid genomes with ONT reads.

If you want an answer to your assembly question you'll have to provide many more details.