search

mikolmogorov / Flye

De novo assembler for single molecule sequencing reads using repeat graphs
Other
779 stars 168 forks source link

small plasmid from nanopore #307

Closed aspitaleri closed 3 years ago

aspitaleri commented 4 years ago

Hi all I have a fastq.gz from a nanopore and I'd like to reconstruct a linear synthetic plasmid about 4k long. I have check the fastq.gz and I got a mean read length of 2,240 bp, median of 1,794 bp and Read length N50 of 3,309 bp. Said so, I am using flye 2.8.1-b1676 version with different options and I am getting different results:

  1. --meta --plasmids out: assembly.fasta 78 contigs, going from 200bp to 200k (only one 4k)

  2. -g 4k --plasmids out: assembly.fasta 3 contigs, 600/724/955 bp

  3. -g 10k --plasmids out: assembly.fasta 1 contig 742 bp

  4. --plamids out: assembly.fasta 2 contigs 472/554/1327 bp

  5. nooptions out: assembly.fasta 1 contig 647bp

First, why do I get so different results, with shorter length with respect to the raw data? Then, do I am doing right for a so small plasmid? I have a fasta template similar to what I should get from the assembly. Could this be used? Thanks for any help!

mikolmogorov commented 4 years ago

Hi @aspitaleri

The parameter set 1 should give you the most comprehensive assembly. You can use your template to check if there is anything similar in the reconstructed contigs.

Do you expect to have only the plasmidic sequence in your reads? It is possible that some other sequences (either biological or artificial) are also present in the data. It is hard to tell why the option sets 2-5 are producing different results, but they are not optimal for metagenomic assembly and it is possible that they are picking up and outputting some junk sequences.

Hope this helps, Mikhail

aspitaleri commented 4 years ago

Hi @fenderglass thanks a lot for your reply. Actually the nanopore run should contain only one single plasmid but the wet people reused an old flow cell, so probably it could contamination indeed. It is not a metagenomic sample but it sounds like. I can use a fasta template to feed the flye run? If so, how do it? Best

mikolmogorov commented 4 years ago

@aspitaleri You can not use fasta template for assembly. But I would try to search the template in the assembled contigs to see if there is anything similar.

mikolmogorov commented 3 years ago

Closing due to inactivity - feel free to reopen if you have any follow-ups.