Dear all, As a part of my project, I had to assemble whole genome reads of a bacterial species in to contigs.
Unfortunately I had no access to Linux server to run algorithms like Velvet or Abyss.
CLC genomic workbench did not work on my system.
I decided to search for online tools, but even in Galaxy I did not find any algorithm except for Trinity which is a RNASeq assembly program.
As a test I used Trinity to assemble my DNASeq reads.
It produced 80 contigs . When I checked these contigs for Multi Locus Sequence Typing or Drug resistance typing which both are on the basis on alignment, I obtained 100% perfect match.
This somehow assured me that Trinity has assembled my reads correctly.
Now my question is, can I rely on the Trinity results?
I am looking forward your guides
Regards
Nazanin
•
written
by
nazaninhoseinkhan
The problem you can face is that Trinity is a transcriptomic assembler that handles the repetitive sequence in a different manner that a genome assembler
For a genomic assembler, reads that are over-represented means repetitive sequences that need to be placed in the genome context
For a transcriptome assembler, over-represented reads simply mean that a determined gene is expressing their transcript at high levels. And try to manage that in a different way
In any case, Trinity will eventually assemble some of your genes, and it will do it right. But it has the tendency to make short fasta files, each of them corresponding to a single gene, whereas the tendency of a genome assembler is to produce long contigs
This is sort of simplified way to explain the differences between the two types of assembler, but hopefully I can give you some important clues about the differences between the two
Dear Antonio Thank you very much. I concluded from your explanation that I can use Trinity for DNA assembly, because my goal is to find genes, not to get a complete assembled version of genome in fasta format. I hope everything go well with you Regards Nazanin
Log in to answer this question.
