Qontas

QONTAS (Quantification of ONT Amplicon Sequences) is a tool designed to take amplicon sequencing reads and generate a clean table of unique variants, their counts, and relative abundances. It uses a "Cookie Cutter" approach to bin, clip, and denoise data based on specific genomic regions of interest.

• Per-Region Length Filtering: Filters reads based on their original length before clipping, allowing for diverse amplicon panels.
• Direct Variant Calling: Extracts standard variant notation (insertions, deletions, and SNVs) directly from the CIGAR and MD tags of the alignments.
• Fast One-Pass Parsing: Bypasses intermediate clustering algorithms and temporary file generation by tallying variants directly from the BAM file in memory.
• Multi-Platform Support: Optimized alignment presets for Oxford Nanopore (ont), PacBio (pb), and Illumina (illumina) sequencing data.

git clone https://github.com/cazzlewazzle89/Qontas.git ./QONTAS
chmod +x ./QONTAS/qontas.py

mamba create -n qontas -c bioconda -c conda-forge minimap2 samtools pysam pandas
conda activate qontas

QONTAS is run using the qontas.py script. It supports multi-region amplicon panels via a single BED file.

The BED file should be tab-separated. Columns 5 and 6 are used to define the expected length of the original read for that specific target. For example:

#Chrom   Start   End     Name    MinLen   MaxLen
chr1     1658    1717    geneA    600      650
chr1     2500    3100    geneB    1100     1250

python qontas.py -i sample.fastq.gz -r ref.fa -b regions.bed -s SampleName -p ont -o Qontas_Out

Because Qontas requires reads to fully span the target coordinates defined in your BED file, paired-end Illumina reads must be merged prior to running the pipeline.

We highly recommend using fastp for this preprocessing step, as it handles both adapter trimming and paired-end merging simultaneously.

Example fastp merging command:

fastp -i R1.fastq.gz -I R2.fastq.gz -m --merged_out sample_merged.fastq.gz

BED Region Length Constraint: When using the -p illumina flag, ensure that the genomic regions defined in your BED file are shorter than your expected merged read length. If the BED coordinates exceed the length of the merged read, the read will be successfully mapped but subsequently discarded by Qontas for failing to span the full target window.

For each sample, QONTAS generates a dedicated folder containing a TSV file for each region defined in the BED file:

Qontas_Out/
└── SampleName/
    ├── geneA.tsv   # TSV with Variant_Name, Count, and Relative Abundance
    └── geneB.tsv   # TSV with Variant_Name, Count, and Relative Abundance

QONTAS includes a utility script qontas_merge.py to combine all individual sample/region TSVs into a single feature table.

Command Line Usage: Simply point the script at your main Qontas output directory (the folder you specified with -o in the main script):

python qontas_merge.py -d Qontas_Out/ -o qontas_table.tsv

Output Format: The script produces a Long Format (Tidy) table. This structure is ideal for downstream analysis and plotting.

• Multi-Sample: Script to merge individual .tsv files into a single master feature table.