MinkeMap

---

Overview

MinkeMap is a command-line tool designed to map sequencing data (FASTQ/FASTA) against a reference genome and visualize coverage and identity statistics in a concentric ring layout. It is heavily inspired by BRIG (BLAST Ring Image Generator).

Key Features

• Fast Mapping: Uses mappy (minimap2) for rapid alignment of Reads or Assemblies.
• Dynamic Scaling: Automatically adjusts track width and gaps to fit the canvas, whether you have 3 tracks or 30.
• Rich Annotations: Supports GenBank (.gbk) features, custom user annotations, and highlight "wedges".
• Focused on Visuals: Exports to .png, .svg, and .pdf with customizable DPI and palettes.
• Data Transparency: Automatically generates a summary.csv and detailed .bed coverage files for every run.

Installation

Mappy Dependencies

MinkeMap relies on mappy (a Python wrapper for minimap2), which requires a C compiler and zlib development headers to build. If you encounter errors during installation, ensure these are installed:

Ubuntu/Debian

sudo apt-get install build-essential zlib1g-dev

CentOS/Fedora

sudo yum install gcc zlib-devel

MinkeMap Installation

From pip

pip install minkemap

From bioconda

conda install -c bioconda minkemap

From source

git clone https://github.com/yourusername/minkemap.git
cd minkemap
pip install .

Quick Start

1. Basic usage with a GenBank reference and a few samples:

minkemap \
  -r reference.gbk \
  -i sample1.fastq sample2.fasta \
  --gc-skew \
  --outdir results

2. Advanced usage with a manifest for FASTQ files and custom styling

minkemap \
  -r reference.gbk \
  -f manifest.csv \
  --annotations custom_regions.csv \
  --highlights background_wedges.csv \
  --palette viridis \
  --min-identity 90 \
  --min-coverage 50 \
  --output my_figure.svg

Input Formats

1. The Manifest File (-f)

Only FASTA files can be read in to MinkeMap with -i. For FASTQ files, a csv that lists a desired name for the track, input files (FASTA or FASTQ), and the type of file is required.

sample,read1,read2,type
SampleA,data/s1_R1.fq.gz,data/s1_R2.fq.gz,illumina
SampleB,data/s2.fasta,,fasta
SampleC,data/s3.fastq,,nanopore

Supported types: illumina, nanopore, pacbio, hifi, fasta, assembly.

2. Custom Annotations (--annotations)

An annotaitons file can be used to draw specific regions of interest on the outermost ring of the plot.

An annotations file is a csv file that lists the reference, start position in relation to the reference, stop position in relation to the reference, a label (optional), and color (optional).

reference,start,stop,label,color
3,15000,22000,Coolio region,#eb5c4b
3,35000,38000,Other coolio region,
3,45000,48000,,#357cb0

3. Highlight Wedges (--highlights)

A file for highlights can be supplied. These are semi-transparent "pizza slices" behind all tracks to highlight regions.

start,end,color,label
10000,20000,#ffcccc,Important Region
45000,48000,#ccffcc,Another Region

Command Line Options

Input / Output

• -r, --reference: Reference genome (GenBank .gbk required for gene arrows, or .fasta).
• -i, --input: List of input FASTA files.
• -f, --input-file: Path to manifest CSV.
• -o, --output: Output filename (default: minkemap_plot.png). Supports .svg and .pdf.
• --outdir: Directory to save images, logs, and summaries (default: minkemap_results).

Aesthetics

• --palette: Color scheme. Accepts presets (whale, viridis, plasma, etc.) or comma-separated hex codes (#ff0000,#0000ff).
• --track-width: Width of each ring (default: 6). Auto-scales if space is low.
• --track-gap: Gap between rings (default: 4). Auto-scales if space is low.
• --dpi: Resolution for images (default: 300).
• --no-backbone: Hides the central black reference axis.
• --no-legend: Hides the legends.
• --label-size: Font size for gene labels (default: 6).
• --title: Add a title to the top of the plot.

Features

• --gc-skew: Adds a GC Skew track (Blue+/Orange-) in the center.
• --annotations: Path to custom annotations CSV.
• --highlights: Path to background highlights CSV.
• --exclude-genes: Comma-separated list of keywords to hide from the gene track (e.g., hypothetical,putative).

Filters

• --min-identity: Minimum % identity (0-100) to display an alignment block.
• --min-coverage: Minimum % query coverage (0-100) required to include a read/contig.

Outputs

Check your output directory for:

1. minkemap_plot.png: The circular visualization.
2. summary.csv: Detailed metrics for every track (Coverage %, Average Depth, Gene Counts).
3. *.bed / *.coverage.csv: Raw data files for every track plotted.
4. minkemap.log: Full execution log for reproducibility.

Note: While MinkeMap streamlines the circular plotting process, it does not aim to replicate the exhaustive customization options of libraries like PyCirclize or BRIG. Therefore, MinkeMap exports all calculated metrics (coverage, identity, skew) as raw data files (CSV/BED), allowing for downstream workflows or other visualization platforms.

Gallery & Examples

Description	Command Used	Output
Standard visualization of Nanopore reads mapped against a FASTA reference.	minkemap -r tests/data/3670018.fasta -i tests/data/*fasta
Standard visualization of Nanopore reads mapped against a GenBank reference.	minkemap -r tests/data/3670018.gbk -i tests/data/*fasta
Adding a GC Skew track (center) and filtering to the rings with >90% coverage.	minkemap -r ./tests/data/3670018.gbk -i tests/data/*fasta --min-coverage 90 --gc-skew
Removing the black reference backbone for a cleaner look, applying the viridis palette, and hiding "hypothetical" proteins from the gene track.	minkemap -r ./tests/data/3670018.gbk -i tests/data/*fasta --no-backbone --palette viridis --exclude-genes "hypothetical"
Uses a CSV Manifest to map multiple samples simultaneously and a Highlights file to mark regions of interest (e.g., AMR genes).	minkemap -r ./tests/data/3670018.gbk -f tests/data/manifest.csv --highlights tests/data/highlights.csv

Why "MinkeMap"

The visual identity of MinkeMap draws inspiration from a local Salt Lake City landmark: the 'Out of the Blue' whale sculpture situated at the 9th and 9th roundabout. There is a natural geometric parallel between the circular traffic intersection and the concentric rings typical of genomic visualizations (such as BRIG plots). MinkeMap leverages this connection, mirroring the roundabout's form and the sculpture's vibrancy to bring a unique aesthetic to circular genome analysis.

And, yes, we are aware the sculpture is of a Humpback whale. We simply couldn't resist the alliteration of Minke with Minimap2, the alignment tool powering this package.

AI Acknowledgement

MinkeMap was developed with the assistance of Google's Gemini for code refactoring, test suite generation, and CI/CD configuration. All AI-generated code has been reviewed, tested, and validated by the authors.

License

Distributed under the MIT License. See LICENSE for more information.