Trimming with custom sequences

[sarah-buddle]

Issue Report

I have run dorado demux and I've noticed that some primer/adapter sequences are left at the start the reads, so I have tried to run dorado trim to remove them. I've tried providing the sequence of the primers using --primer-sequences and also without this parameter. In both cases, some reads are trimmed but I find that in the output file many reads (100s out of 3195 total reads) still contain these sequences at the start. Is there a setting I should change so the trimming works for all the reads?

Steps to reproduce the issue:

The command I am using is:
dorado trim input.fastq --emit-fastq --primer-sequences primers.fasta > output.fastq

The input file was the output of the dorado demux command (with trimming enabled) run on a previously basecalled fastq file.

Custom primer sequences that are still found in output:

>seq1
TTCAGACGTGTGCTCTTCCGATCT
>seq2
CCTACACGACGCTCTTCCGATCT
>seq3
CCTACACGACG

Run environment:

• Dorado version: 0.8.3+98456f7
• Dorado command: see above
• Operating system: CentOS Linux 7 (Core)
• Hardware (CPUs, Memory, GPUs): CPUs, 10G memory currently (I'm running a small file for testing)
• Source data type: fastq, previously basecalled from pod5
• Source data location (on device or networked drive - NFS, etc.): On HPC
• Details about data (flow cell, kit, read lengths, number of reads, total dataset size in MB/GB/TB): Test fastq file is 1.3M

Logs

[2024-12-13 16:18:57.672] [info] Running: "trim" "input.fastq" "--primer-sequences" "primers.fasta" "--emit-fastq" "-v"
[2024-12-13 16:18:57.672] [debug] > adapter/primer trimming threads 58, writer threads 6
[2024-12-13 16:18:57.673] [info] - Note: FASTQ output is not recommended as not all data can be preserved.
[2024-12-13 16:18:57.677] [info] > starting adapter/primer trimming
[2024-12-13 16:18:57.727] [debug] Total reads processed: 3195
[2024-12-13 16:18:57.781] [info] > Simplex reads basecalled: 3195
[2024-12-13 16:18:57.781] [info] > finished adapter/primer trimming

[malton-ont]

Hi @sarah-buddle,

There's nothing obvious in your command that should stop this from working. Are you able to share the test data file with us?