Assembly of oxford nanopore sequencing data

Question

[MG_19]

From the sequencing run multiple fastq files are generated after basecalling from fast5 files for a particular barcode of a species. Should I merge them before starting further analysis?

Answer 1

[WouterDeCoster]

You didn't tell us what the next step would be in your analysis, so it's a little hard to give you a full answer. You can definitely merge them, but you probably don't have to. Most tools (aligners and assemblers) can take multiple files at the same time as input.

[MG_19]

Thank you for your suggestion. Actually I will further process the data in this way like quality check, trimming, assembly & annotation. I have found few tools which take multiple files as input.

Answer 2

[H.Hasani]

That was also my question; I'm trying to evaluate my first ONT data and I did both,

• use the fastq files directly, align them to the reference genome using guppy then merge the bam files (used both samtools and bam_convert within guppy suit).
• merge the fastqs then align that file using both minimap2 & guppy.

The number of mapped reads either way remaind identical. However, the merging bam files produced different output on igv compared to merging fastq first. The later looked way more reasonable and fitted better when considering all the produced alignment statistics. Therefore, it looks to me that merging the fastq files first is the way to go.

How did you at the end proceed?