if i do mistake in library type parameter of trinity assembly, do i need to reassemble it again?

Question

[biologo]

Dear friends,

i was doing the trinity assembly, but i made a small mistake that for the --library type parameter, i used the FR, but actually it was RF, till i did the blast that i found the known gene matched to the reverse strand, cause the trinity take times and CPU, can i just reverse the Trinity.fasta file, and no need redo it again? does it works???

thank you so much for your help.

[colindaven]

No, you'll need to redo it, but you should get a much better assembly.

Colin

[biologo]

i got it, thank you, but your extral advice remind me that actually the quality is not good, even i isolate the llongest unigene, i got over 430,000 transcriptome.

[colindaven]

Thats a lot of transcripts, and they are probably highly fragmented. I'm sure you did read trimming before, right ? Check your data with FASTQC before and after trimming too.

[biologo]

yes, i did. trim the adaptor and filter some short or low quality reads, but the trinity result seems no better options, thank you for your patience and your kind reply.

[lakhujanivijay]

You might try out Bridger

[biologo]

yes, i also considered that, but my colleague suggested me use ingap-cdg, have you ever use that, he told me it really works, and i also did the test, and only left 7,000,000 transcripts, seems good, but i am still doubt about the result.

Answer 1

[colindaven]

It might be low quality data, but please tell us step by step what was done exactly. Also, you can run it through a trinity workflow for example on Galaxy main.

How many reads do you have ?

Another assembler - CLC is easy and quite good, or soapdenovo-trans has a decent reputation - might help too.

Also, just map your transcripts to the genome (I like Gmap) then visualize with Jbrowse or IGV etc. Are multiple redundant transcript fragments present ?

[biologo]

accepted,but the thing is there's no reference genome for this species, and for some reasons, the genome assembly is really hard, to much AT ratio and repeat region, so, we just wanna focusing on the transcriptome assembly. thank you for your kind reply.