HAlign-4

HAlign 4: A New Strategy for Rapidly Aligning Millions of Sequences.

Documentation:

• Detailed usage & examples: docs/usage.md
• Source install & dependencies: docs/install.md
• Tests: docs/test.md

---

Install (Conda)

Conda is the recommended installation method for end users.

conda install -c malab halign4

Verify:

halign4 --version
halign4 -h

Source installation: see docs/install.md.

---

Quick start

The repository includes small datasets under test/data/ which are perfect for a first run.

Minimal example:

halign4 \
  -i test/data/mt1x.fasta.gz \
  -o mt1x.out.fasta

---

Parameters (overview)

The most important parameters are:

• -i/--input: input FASTA (required)
• -o/--output: output aligned FASTA (required)
• -w/--workdir: working directory (optional; default: ./tmp-<random>)
• -p/--msa-cmd: MSA method (keyword: minipoa/mafft/clustalo, or a custom template)
• -c/--center-path: provide a reference/center FASTA (optional)
• --keep-first-length / --keep-all-length: keep reference length coordinate rules

For the full parameter list and detailed examples, see docs/usage.md.

---

Tests

See docs/test.md for how to run tests under the test/ directory.

---

Citation

If you use HAlign-4 in academic work, please cite:

HAlign 4: a new strategy for rapidly aligning millions of sequences. Bioinformatics, 2024, 40(12): btae718. https://doi.org/10.1093/bioinformatics/btae718