OmniAge

A comprehensive Python framework for aging research and biological clock calculation

Authors:

• Zhaozhen Du (Shanghai Institute of Nutrition and Health, CAS) - duzhaozhen2022@sinh.ac.cn
• Andrew E. Teschendorff (Shanghai Institute of Nutrition and Health, CAS) - andrew@sinh.ac.cn

📖 Overview

OmniAge offers high-level interfaces for calculating a vast suite of aging-related clocks and biomarkers. It is designed to integrate seamlessly with standard bioinformatics workflows (e.g., scanpy, pandas).

The package covers three main categories:

1. Epigenetic (DNAm) Clocks: Including Chronological, Biological (e.g., GrimAge, DunedinPACE), Cellular (Mitotic), and Gestational age clocks.
2. Transcriptomic (RNA) Clocks: Includes clocks based on bulk or single-cell RNA-seq data (e.g., scImmuAging, PASTA).
3. Proteomic Clocks Comprises age predictors based on plasma protein expression levels(e.g., WyssCorayOrganAge, GladyshevOrganAge).
4. Other DNAm Human Aging-Related Surrogate Biomarkers Provides specialized epigenetic markers for plasma proteins, inflammation, clonal hematopoiesis of indeterminate potential (CHIP), lifestyle traits, and specific disease risk indices (e.g., Episcores, CRP, DNMT3A, McCartney_Trait, and SmokeIndex).
5. Other Species: Integrates universal cross-species clocks alongside models tailored specifically for non-human species(e.g., UniversalPanMammalianClocks, and MeerMouse).

✨ For a complete list of supported clocks and detailed usage examples, please refer to our Tutorial Notebook.

🛠 Installation

Install via pip with the subdirectory flag:

git clone https://github.com/Duzhaozhen/OmniAgePy.git
pip install .

[!TIP] If you encounter errors building pyarrow or h5py (common on Linux servers), we recommend pre-installing these dependencies via Conda:

conda install -c conda-forge pyarrow h5py