Polishing Nanopore data

Question

[nbargues]

Hi, I have Nanopore data from sequencing of a specific human gene . I have the equivalent with Illumina.

I have made a pipeline using only ONT data for assembling (minimap2) and calling variant (longshot) but, with this specific gene, I can't found a specific SNP that I know is present via wet lab.

I need to improve the quality of my data my correcting my read.

So the question are : Do you have knowledge of a software that correct or polish Nanopore data ( with or without need of Illumina read ) before or after alignment ?

Thanks

[GenoMax]

See: https://github.com/nanoporetech/ont-assembly-polish
https://www.biorxiv.org/content/10.1101/2019.12.17.864991v1.full.pdf

[nbargues]

I saw that approach but sadly I have not enough memory for running canu :/

[GenoMax]

Have you tried to extract the reads in the region of interest (from minimap2 alignment)? Not sure if you can polish that subset?

Perhaps try something simple as a multiple sequence alignment (sounds like you have a good reference). You may need to downsample if you have a ton of reads. If the SNP is present in your data you should be able to see it there. Do you see the SNP in your Illumina data?

[colindaven]

Bit late, but is this data from a specific gene a) amplicons ? or b) cDNA sequencing etc ? What is your expected coverage ?

Generally for nanopore polishing I would recommend Racon. You can probably use long reads to correct the long reads (Racon) but also your shorter illumina reads if neccessary.

Its not clear if you have an assembly or just an alignment so far. As genomax says you can extract the specific reads, or a subset, and just correct them.