Trimming nanopore reads

Question

[Nodilan]

Good morning ,

Can I use trimmomatic in order to trim my reads generated using nanopore technology ?

thanks,

Answer 1

[colindaven]

Yes, but it probably will not work well, as nanopore read trimming was not Trimmomatics' intended data type.

Have a look at Fastp, chopper or the Porechop_ABI fork of Porechop https://github.com/bonsai-team/Porechop_ABI

Seqtk is another possibility.

[dalibenam64]

Hello, -Can I skip trimming my reads since I'm filtering them by taking those with the highest scores using fastp? I'm performing variant calling analysis on these reads to detect specific SNPs, insertions, and indels (I already know the positions of these mutations).

-I'm aligning my reads to reference gene sequences and I'm working on fungus.

-basecalling was performed with Guppy

-Just for the record, when I compare my mutation frequency (generated using fastq.gz file not trimmed)t o those in the literature, I find the same result.

thank you in advance !!

[GenoMax]

Can I skip trimming my reads since I'm filtering them by taking those with the highest scores using fastp?

While you could maybe you should not. There may be extraneous sequence that has good scores, which would would not want, IF you were going to do any de novo work.

I'm aligning my reads to reference gene sequences and I'm working on fungus.

In this case you could skip trimming since the aligner should soft clip parts of reads that do not map. There is a small danger that some reference sequences at NCBI may contain adapter contamination (people are not always careful in submissions).

when I compare my mutation frequency (generated using fastq.gz file not trimmed)

As long as you have ample evidence (multiple reads supporting the SNP) it should be fine.

[dalibenam64]

thank you very much I'm working on cds sequence of the gene of interest when I run this command :

  fastp -i path/input_folder/barcode13-Cytb.fastq.gz -o path/output_folder/barcode13-Cytb.fastq.gz -A --qualified_quality_phred 10 --thread 16

=> it works

but now when I use this :

   fastplong -i path/input_folder/barcode13-Cytb.fastq.gz -o path/output_folder/barcode13-Cytb.fastq.gz -A --qualified_quality_phred 10 --thread 16

I have this error that I can't understand

"Before filtering:
total reads: 52694
total bases: 83676141
Q20 bases: 40084516(47.9044%)
Q30 bases: 17434998(20.8363%)

After filtering:
total reads: 0
total bases: 0
Q20 bases: 0(-nan%)
Q30 bases: 0(-nan%)

Filtering result:
reads passed filter: 0
reads failed due to low quality: 0
reads failed due to too many N: 0
reads failed due to too short: 0
terminate called after throwing an instance of 'std::bad_array_new_length'
  what():  std::bad_array_new_length
Aborted (core dumped)
Error with fastp on barcode13-Cytb.fastq.gz"

[GenoMax]

terminate called after throwing an instance of 'std::bad_array_new_length'

It is possible that you may have identified a bug with fastplong. What is the length of your longest read? You can create a bug report here: https://github.com/OpenGene/fastplong/issues

[GenoMax]

To complement the discussion here: fastplong (LINK) is a version of fastp meant for long reads such as ONT/PacBio.

Answer 2

[cfos4698]

I recommend using nanoq or filtlong for quality/length filtering/trimming of nanopore reads

[Nodilan]

can I still use fastqc and multiqc to check the quality of my nanopore reads ?

[GenoMax]

While you can, you will want to use a package meant for long reads like PycoQC. It requires the sequencing_summary file from the nanopore run.

[colindaven]

You can, but I'd use Fastp and Multiqc instead. Faster and better output of %Q20 and %Q30 reads for ONT (in my opinion). Also PycoQC and or the Nanoplot / Nanopack toolset https://github.com/wdecoster/nanopack might be useful.

[zau saa]

This page may help https://help.nanoporetech.com/en/articles/6628901-should-i-qc-my-data-and-what-tools-to-use