Entering edit mode
5.2 years ago
binlu1981
•
0
The 3rd generation sequencing tech can help us obtain full-length transcriptome. I want to know if it's much better than 2nd sequencing transcriptome when help annotate genome assembly? In terms of completeness, especially accuracy?
I am working on annotation of a complex genome assembly. I don't know if the full-length transcriptome is much better for annotation than 2nd generation RNA-seq. The price of the former is much higher than later one. My goal is to get better annotations without wasting money.
I will be appreciate if any suggestions can be provided.
It depends on the type of project (organism, goal, budget, etc.). NGS RNA-seq is still very useful for gene discovery, particularly for species being sequenced for the first time. It is cheap and gives you lots of data. On the other hand, 3rd Gen Sequencing can be more useful for accurately discovering certain types of gene variants, e.g. different isoforms.
Thanks for reply. Agree with your opinion.
Whe have seen in complex genomes that both are complementary. For genomes with few introns (e.g Fungi) no need to use long reads.
Thanks. I got it for simple genome, but for complex genome, I don't quite understand. The long-read sequencing tech can get full-length transcripts. Should it be more comprehensive and accurate? Why we still need short-read transcriptome for complex genome annotation?
I mixing up projects. We got a project where both were complementary because they just came from different samples from different tissues. So not really related to the technology here... sorry. I got another project where we got the first generation pacbio transcriptome data, and the transcripts were really bad containing many errors. So mapping them to the genome didn't provide good results but helped to anchor abinito prediction (evidence-driven). The Illumina data were much more informative.
Got it. Did you correct the sequencing errors using short read? Or after correction, there are still a lot of errors? Did you try Pb CCS mode which should decrease the sequencing error?
No I didn't correct anything, it was just a test few years ago. They have a lot improved. We used recently new pacbio transcriptome data (CCS) and they map very well to the genome.