AssertionError: DNA sequence must be a nucleotide sequence of ATGCRYSWKMBDHVN

Hello,

Thanks for the wonderful tool!

It seems to run fine on many datasets, but for me it fails for one specific dataset, with the following error report:

Activating conda environment: .snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_
WARNING:root:Output directory results/F6D39/NanoMotif/ already exists
Traceback (most recent call last):
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/bin/nanomotif", line 10, in <module>
    sys.exit(main())
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/main.py", line 514, in main
    motif_discovery(args)
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/main.py", line 299, in motif_discovery
    assembly = nm.load_assembly(args.assembly)
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/dataload.py", line 46, in load_assembly
    return Assembly(load_fasta(path))
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/seq.py", line 13, in __init__
    self.assembly = {name:DNAsequence(seq) for name, seq in assembly.items()}
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/seq.py", line 13, in <dictcomp>
    self.assembly = {name:DNAsequence(seq) for name, seq in assembly.items()}
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/seq.py", line 54, in __init__
    self._check_sequence(sequence)
  File "/scratch/brbloemen/FLUPOUL/FLUPOULmeta/workflow/.snakemake/conda/1c99e828952b31ca3eef31b2ad43ef5c_/lib/python3.9/site-packages/nanomotif/seq.py", line 71, in _check_sequence
    assert all(letter in self.iupac for letter in list(set(sequence))), f"DNA sequence must be a nucleotide sequence of {''.join(self.iupac)}"
AssertionError: DNA sequence must be a nucleotide sequence of ATGCRYSWKMBDHVN

The fastq files with methylation data were created as follows with dorado v0.7.0:

model=/scratch/brbloemen/dorado_models/dna_r10.4.1_e8.2_400bps_sup@v5.0.0
dorado basecaller $model "$INPUT_POD5_PATH" --device cuda:all -b 320 --modified-bases 4mC_5mC 6mA > "basecalling_dor7mod5_modified/${filename}_mods.bam"

samtools fastq "basecalling_dor7mod5_modified/${filename}_mods.bam" -T '*' | pigz -p9 > "basecalling_dor7mod5_modified/${filename}_mods.fastq.gz"

For the assembly I used Flye v2.9.3, using the --meta and --nano-hq parameters.

I aligned the fastq to the assembly fasta with minimap 2.26:

minimap2 -ax map-ont -y -t {threads} {input.assembly_fasta} {input.fastq} > {output.sam}

and filtered, sorted and indexed with samtools:

samtools view -hb -F0x904 -@{threads} {input.sam} | samtools sort -@{threads} > {output.bam}
samtools index -@{threads} {output.bam} -o {output.bai}
samtools faidx {input.assembly} -o {output.fai}

MicrobialDarkMatter / nanomotif

AssertionError: DNA sequence must be a nucleotide sequence of ATGCRYSWKMBDHVN #56