How to assemble viral genome using MIRA assembler

Question

[Kumar]

Hi, I am looking to make hybrid assembly of a viral genome. I have got (R1 and R2) and a merged file fastq.gz, generated from Nanopore. I used SPAdes but it makes fragmented assembly. Please let me know if MIRA assembler enter link description here works for hybrid assembly with Nanopore sequencer.

Answer 1

[Istvan Albert]

There are quite a few assemblers designed to be used with Nanopore data: Canu, Flye, Falcon etc.

I had good experiences with those. As far as I know, mira is not recommended for nanopore. The documentation says:

Do not use MIRA if you have PacBio or Oxford Nanopore reads.

By the way here is a paper published today on the topic, a new assembler:

Efficient assembly of nanopore reads via highly accurate and intact error correction

• https://www.nature.com/articles/s41467-020-20236-7

[Kumar]

Hi,

Thank you for your reply. I have both Illumina and Nanopore reads and am looking to make a hybrid assembly. My implication is that if I use Canu, Flye or Falcon, are these appropriate for making hybrid assembly of viral genomes? I tried Unicycler but it is designed for bacterial specifically.

Since the samples were collected from an animal host, do I need to remove the host sequence from the reads before making assembly. I would appreciate your suggestion!

[Istvan Albert]

there are different strategies one might use, depending on how much of each data is available

the "modern" approach is to assemble the Nanopore reads then "polish" the resulting contigs using the Illumina reads.

https://github.com/nanoporetech/ont-assembly-polish

if you can identify the host contamination then yes it would be best to remove those reads.