omicsync

A Python library for multi-omics data harmonisation.

omicsync handles the tedious work of aligning sample IDs, normalising each modality consistently, and exporting to downstream tools so you can focus on biology, not data wrangling.

Installation

pip install omicsync

With optional extras:

pip install "omicsync[mofa]"       # MOFA2 factor analysis
pip install "omicsync[geo]"        # GEO data loading
pip install "omicsync[anndata]"    # AnnData export
pip install "omicsync[torch]"      # PyTorch tensor export
pip install "omicsync[all]"        # Everything

Quick Start

import omicsync as oms
from omicsync.loaders.csv import load_multimodal_csv

# Load multiple modalities from CSV files
dataset = load_multimodal_csv({
    "rna":     "brca_rna.tsv",
    "protein": "brca_rppa.tsv",
    "cnv":     "brca_cnv.tsv",
}, study_id="TCGA-BRCA")

# Align, normalise, filter — all chainable
dataset.align_samples().normalize().filter_features(min_variance=0.01)

# Export to DataFrame or MOFA2
df = dataset.to_dataframe()          # samples × features, prefixed columns
mofa_input = dataset.to_mofa2()      # dict ready for mofapy2 entry_point

Features

• Sample harmonisation — TCGA barcode parsing, fuzzy ID matching, coverage reporting
• Per-modality normalisation — auto-detection of count/TPM/M-value formats
• Chainable API — dataset.align().normalize().filter_features()
• sklearn compatibility — use OmicsSyncTransformer in a Pipeline
• Multiple export formats — DataFrame, dict, MOFA2, PyTorch tensor, AnnData
• Open Targets integration — query target-disease associations via GraphQL
• Type hints throughout — fully typed public API

Supported Data Sources

Supported Modalities

Documentation

• Quickstart guide
• API reference
• Tutorial: TCGA BRCA
• Tutorial: Custom CSV data

Citation

If you use omicsync in your research, please cite:

Paterson V. (2026). omicsync: A Python library for multi-omics data harmonisation. GitHub: github.com/vi-c-ky/omicsync

Contributing

Contributions are welcome. Please open an issue or pull request on GitHub.

1. Fork the repository
2. Create a feature branch (git checkout -b feature/my-feature)
3. Write tests for new functionality
4. Run the test suite (pytest tests/)
5. Open a pull request

License

MIT — see LICENSE for details.