GitHub

SeededAlignment

Documentation for SeededAlignment.

Introduction

SeededAlignment.jl is a flexible high-performance Julia package for aligning DNA sequences. In particular, it can be used to create alignments that preserves reading frame in coding regions. The package provides tools for:

  • High-performance pairwise alignments using k-mer seeds and chaining
  • Producing reading frame preserved multiple sequence alignments
  • Customizable alignment operations and scoring systems
  • Cleaning single indel noise from pairwise codon alignment

This documentation provides a deeper look into the package’s functionality, design, and customization options.

If you're looking for installation, check out the README on GitHub.

When to Use SeededAlignment.jl

This package is ideal for:

  • Aligning coding DNA sequences where preserving the reading frame is critical
  • Performing fast pairwise alignments using seed-and-extend strategies
  • Creating codon-aware multiple sequence alignments with a trusted reference
  • Customizing alignment scoring to handle mismatches, gaps, or biological constraints

Core Features

  • 🔗 Seeded Pairwise Alignment — Fast alignment using k-mer seeding and chaining
  • 🧬 Codon-Aware MSA — Align multiple sequences while preserving codon structure
  • ⚙️ Custom Move Sets — Define allowed alignment operations (e.g., frameshifts, gaps)
  • 🧠 Flexible Scoring — Easily adjust match/mismatch/gap penalties
  • 🧹 clean reading frame - Remove indels which break reading frame for a pairwise alignment
  • 📄 FASTA Support — Simple reading and writing of sequence data

Usage Example

Here's a full example using standard codon-aware alignment tools:

using SeededAlignment
using BioSequences

# Create a reference and target sequences
ref = LongDNA{4}("ATGACGTGA")  # Reference sequence
seqs = [LongDNA{4}("ATGTCGTGA"), LongDNA{4}("ATGACGAGA")]

# Use built-in codon_alignment parameter settings
moveset = std_codon_moveset()
scoring = std_codon_scoring()

# Align sequences to the reference
alignment = msa_codon_align(ref, seqs, moveset, scoring)

# Save result to FASTA
write_fasta("alignment.fasta", alignment)