CORAL

Comparative Orthologous Read-based Analysis of Lineage Substitutions

CORAL is a tool for scalable extraction, detection, and analysis of point mutations across species evolutionary history. It aligns multiple species to a shared reference genome, simulates reads, filters alignments by mapping quality, extracts unambiguous trinucleotide substitutions, and summarizes mutation rates and mutation spectra.

---

Reference

Preprint available at https://doi.org/10.64898/2026.02.02.703326

---

Pipeline overview

---

Installation

Requirements

• Linux (or WSL2 for windows)
• Conda (Miniforge or Anaconda recommended)

Recommended installation

git clone https://github.com/asafpinhasitechnion/CORAL.git
cd CORAL
conda env create -f environment.yml
conda activate coral-env
pip install -e .

Verify installation

coral --help
samtools --version
bwa
datasets --version

The provided environment.yml installs all required dependencies, including:

• Python 3.10
• BWA (classic)
• SAMtools
• NCBI Datasets CLI
• unzip
• All required Python dependencies

Optional: PHYLIP (for phylogenetic inference)

PHYLIP is not required for the core pipeline.

Install only if using phylogenetic inference via coral run_multi or coral run_phylip:

conda install -c bioconda phylip

---

Quick start

Three-taxon pipeline (outgroup + two ingroups)

coral run_single \
  --outgroup Saccharomyces_mikatae_IFO_1815 GCF_947241705.1 \
  --species Saccharomyces_paradoxus GCF_002079055.1 \
            Saccharomyces_cerevisiae_S288C GCF_000146045.2 \
  --output ../test_output \
  --mapq 60 \
  --suffix test

This runs the full pipeline, including genome download, reference indexing, read simulation, alignment, mutation extraction, and summary table and plot generation.

---

Multi-species analysis (experimental)

coral run_multi \
  --species-list '[["Drosophila_melanogaster","GCF_000001215.4"],["Drosophila_sechellia","GCF_004382195.2"],["Drosophila_mauritiana","GCF_004382145.1"],["Drosophila_simulans","GCF_016746395.2"]]' \
  --outgroup Drosophila_simulans \
  --output ../test_output \
  --run-id drosophila_test \
  --mapq 60

Note: Multi-species mode is experimental and intended for exploratory analyses.

---

Functional workflow

Step 1: Genome preparation

• Download genomes by NCBI assembly accession
• Index the reference genome for alignment

Step 2: Read simulation and alignment

• Simulate FASTQ reads by sliding a window across genomes
• Align simulated reads to the outgroup reference
• Filter alignments by MAPQ and coverage
• Allow customization of aligner and parameters

Step 3: Mutation detection

• Generate pileups from reference and aligned BAMs
• Extract unambiguous trinucleotide substitutions
• Optionally retain genomic positions

Step 4: Normalization and analysis

• Normalize mutation counts by underlying trinucleotide abundance
• Collapse complementary strands into canonical spectra
• Generate summary tables and visualizations

---

Output overview

Each run produces a self-contained output directory containing:

• Mutations/*_mutations.csv.gz – per-branch mutation lists
• Mutations/*_mutations.json – trinucleotide mutation counts
• Tables/*.tsv – normalized mutation spectra
• Plots/*.png – diagnostic and summary plots

Mutation files are named:

<taxon1>__<taxon2>__<reference>__mutations.*

This indicates mutations inferred on the branch leading to taxon1 since divergence from taxon2, using reference as the outgroup genome.

See OUTPUT_FORMAT.md for full file format and naming conventions.

---

Documentation

• tutorial.ipynb – command-line tutorial and examples
• OUTPUT_FORMAT.md – output file structure and naming conventions

---

Citation

Details, benchmarking, and results are available in the preprint: https://doi.org/10.64898/2026.02.02.703326

The final reference will be updated upon publication.

---