Automated Genome Downloading & Windowing Tool
Part of the MethylSense Package

---

Overview

A versatile Bash script that generates genomic windows of specified sizes from FASTA files. Can also automatically download genomes from Ensembl or UCSC databases. This tool is part of the MethylSense package preprocessing pipeline.

Features

• Direct FASTA input - Process local genome files with --genome flag (v1.1)
• Automatic genome downloading from Ensembl or UCSC databases
• Flexible window sizes with sensible defaults
• macOS compatible - Works on both Linux and macOS (v1.1)
• System tools fallback - Use system bedtools/samtools with --no-micromamba (v1.1)
• Progress tracking with visual progress bars
• Dry-run mode to preview commands before execution

Requirements

• Bash (3.2+, macOS compatible)
• bedtools and samtools (via system install OR micromamba)
• curl (for downloading genomes)

Installation

Quick Install (Recommended)

git clone https://github.com/markusdrag/GenomeToWindows.git
cd GenomeToWindows
chmod +x window_those_genomes.sh

If you have bedtools and samtools installed, you're ready to go with --no-micromamba.

With Micromamba (Optional)

For automatic dependency management:

./install.sh

Usage

Process a Local FASTA File (Recommended)

# Process your local genome file
./window_those_genomes.sh \
  --genome /path/to/genome.fna \
  --windows 200,400,500,750,1000 \
  --output ./my_windows \
  --no-micromamba

Download and Process

# Download human genome from Ensembl and generate default windows
./window_those_genomes.sh --species Homo_sapiens

# Download chicken genome with benchmarking window sizes
./window_those_genomes.sh --species Gallus_gallus --windows 200,400,500,750,1000

# Download mouse genome with custom window sizes
./window_those_genomes.sh --species Mus_musculus --windows 1000,5000,10000

# Use UCSC database instead of Ensembl
./window_those_genomes.sh --species Homo_sapiens --database ucsc

# Preview download commands without executing (dry-run)
./window_those_genomes.sh --species Gallus_gallus --dry-run

Process Directory of Genomes

./window_those_genomes.sh --input ./my_genomes --output ./my_windows

Command-Line Options

Options:
  -g, --genome FILE    Direct path to a FASTA file (.fna/.fa/.fasta)
                       This is the PREFERRED option for local files
  --fai FILE           Direct path to a FAI index file (skips samtools indexing)
  -s, --species SPECIES  Scientific name (e.g., "Homo_sapiens") - downloads genome
  -d, --database DB      Database: ensembl or ucsc (default: ensembl)
  -i, --input DIR        Input directory with .fna/.fa files (default: ./genomes)
  -o, --output DIR       Output directory for BED files (default: ./windowed_genomes)
  -w, --windows SIZES    Comma-separated window sizes in bp
                         Example: -w 200,400,500,750,1000
  --no-micromamba        Use system bedtools/samtools instead of micromamba
  --dry-run              Preview commands without executing
  -h, --help             Show help message
  -v, --version          Show version

Examples

Quick Start: Download and Window a Genome

The easiest way to get started is to let the script automatically download a genome from Ensembl. Just specify the species name and your desired window sizes:

# Download human genome from Ensembl and generate 1kb, 5kb, and 10kb windows
./window_those_genomes.sh --species Homo_sapiens --windows 1000,5000,10000

# Download chicken genome with MethylSense benchmarking sizes (200bp to 1000bp)
./window_those_genomes.sh --species Gallus_gallus --windows 200,400,500,750,1000

# Download mouse genome from UCSC instead of Ensembl
./window_those_genomes.sh --species Mus_musculus --database ucsc

The script will:

1. Download the genome FASTA from the selected database
2. Index it with samtools faidx
3. Generate BED files with non-overlapping windows for each specified size

Using a Local Genome File

If you already have a genome FASTA file on your system, use --genome to process it directly. This is faster and avoids re-downloading:

./window_those_genomes.sh \
  --genome /path/to/Gallus_gallus.GRCg7b.dna.toplevel.fna \
  --windows 200,400,500,750,1000 \
  --output ./benchmark_windows \
  --no-micromamba

Tip: Use --no-micromamba if you have bedtools and samtools already installed on your system.

Skipping Indexing with Pre-existing FAI

If you already have a .fai index file, you can skip the indexing step entirely by providing it with --fai:

./window_those_genomes.sh \
  --genome /path/to/genome.fna \
  --fai /path/to/genome.fna.fai \
  --windows 1000,5000,10000

This is useful when working with very large genomes where indexing can take several minutes.

Supported Species (for download)

Ensembl

All species in Ensembl (100+ species). Common examples:

• Homo_sapiens, Mus_musculus, Rattus_norvegicus
• Danio_rerio, Gallus_gallus, Bos_taurus

UCSC

Pre-configured mapping:

• Homo_sapiens → hg38
• Mus_musculus → mm39
• Gallus_gallus → galGal6
• And more...

Output

BED files in ./windowed_genomes/ (or custom --output):

• {genome_name}_{window_size}bp.bed
• Example: Gallus_gallus_500bp.bed

Changelog

v1.1.0 (2026-01-02)

• NEW: --genome flag for direct FASTA input
• NEW: --fai flag for direct FAI index input
• NEW: --no-micromamba to use system tools
• FIX: macOS compatibility (removed GNU grep -P)
• FIX: Bash 3.x compatibility (removed associative arrays)

v1.0.0

• Initial release

License

MIT License - see LICENSE file for details

Citation

Drag, M., et al. (2025). New high accuracy diagnostics for avian Aspergillus fumigatus
infection using Nanopore methylation sequencing of host cell-free DNA and machine
learning prediction. bioRxiv. https://doi.org/10.1101/2025.04.11.648151

Contact

• Issues: GitHub Issues
• Email: markusdrag@gmail.com