OmicOS: A Comprehensive Omics Ecosystem Infrastructure and Agent System for the AI Era

OmicOS: комплексная инфраструктура экосистемы омиксных технологий и агентная система для эпохи искусственного интеллекта
Zhi Luo, Yuanyan Xiong, Xuegong Zhang, Hao Zhu, Hongwu Du, Liping Zhou, Z H E N G M Jiang, Sheng Mao, Y Shi, Pi Liu, Zehua Zeng, Xu Meng, LH Hu, Chen Li, Xuejiao Ma, Lianchong Gao, Xuelei Wang, Yawen Zheng, Jieshen Xian, Ziheng Lin, Yifan Lu, Wenzhuo Tang, Qiangwei Peng, Yuqing Ma, Cencan Xing
2026-06-16

BiomniBenchOmicOSagentic biologysingle-cell and spatial omicsstate-aware capability contracts
Abstract Biology has accumulated a vast ecosystem of omics methods, but much of this ecosystem remains built for expert humans rather than scientific agents. Methods are scattered across Python packages, R/Bioconductor and CRAN workflows, command-line tools, incompatible data containers and implicit object states, making even routine analyses difficult for an AI system to choose, execute and verify reliably. Here we introduce OmicOS, a comprehensive omics ecosystem infrastructure and agent system that turns OmicVerse V2, an open-source omics community, into an executable foundation for agentic biology. OmicVerse V2 provides the community substrate: scalable AnnDataOOM-compatible rust backends, agent-friendly Python algorithms for single-cell, spatial, bulk and multi-omics analysis, interfaces to single-cell foundation models, and Python-native reconstructions of historically R-centred Bioconductor/CRAN-style workflows. OmicOS makes this substrate actionable by registering analytical functions as state-aware capability contracts, allowing agents to inspect live data objects, select valid methods, execute controlled workflows and record provenance. The result is not a fixed pipeline, but a programmable omics environment in which agents compose real analyses from verified community methods rather than inventing tools. Across external and purpose-built benchmarks, OmicOS ranked first among the evaluated systems, reaching 81.2% on BiomniBench. Adding OmicVerse to a minimal agent improved task completion by up to 34.2 percentage points with qwen-3.6-35b, and controlled ablations showed that the gains came from registry-grounded execution rather than from larger models, documentation retrieval or unrestricted tool exposure. The same infrastructure scaled to atlas-sized data, reproduced R-centred workflows in Python and converted external pathology software into agent-usable skills. In a discovery task starting from a whole-body spatial map and the term “Alzheimer’s disease”, OmicOS composed a non-canonical workflow that integrated spatial expression, genetic association, eQTL and colocalization evidence to nominate a colon epithelial risk axis centred on PICALM, CD2AP and CR1. Together, OmicVerse and OmicOS define an open foundation for AI-era omics, showing how a community of biological methods can be transformed into a reliable, extensible and agent-operable system for discovery. Highlight OmicVerse 2.0 consolidates 694 methods spanning 11 omics domains into agent-callable high-level APIs. RebuildR automatically reconstructs and evolves R/Bioconductor methods as Python-native implementations under output-equivalence gates. OmicOS establishes a state-of-the-art omics agent harness, ranking first on general omics benchmarks across models and substantially improving the analytical capability of local open-source models. Compositional use of ecosystem modules nominates a colon epithelial axis associated with Alzheimer’s disease risk. External algorithm packages supporting automatic iterative evolution can be integrated into the OmicOS ecosystem.
1
Adding OmicVerse to a minimal agent improved task completion by up to 34.2 percentage points with qwen-3.6-35b; ablations attributed gains to registry-grounded execution rather than model size, documentation retrieval, or unrestricted tools.
2
Its state-aware capability contracts enable agents to inspect live data objects, select valid methods, execute controlled analyses, and record provenance.
3
OmicOS ranked first among evaluated systems across external and purpose-built benchmarks, achieving 81.2% on BiomniBench.
4
OmicOS transforms OmicVerse V2 into an executable foundation for agentic biology, integrating scalable backends, agent-friendly algorithms, foundation-model interfaces, and Python-native omics workflows.
5
The infrastructure supports atlas-scale data, reproduces R-centred workflows in Python, and converts external pathology software into agent-usable skills.

OmicOS, an executable omics ecosystem infrastructure and agent system for agentic biology

The capability of an AI-agent-accessible omics environment to select, execute, verify, compose, and reproduce state-aware analytical workflows with provenance across single-cell, spatial, bulk, and multi-omics data

Publication Details
Publication Date
2026-06-16
Journal
Publisher
ISSN
Cited by
1
Access Type
Author Information
Authors
Zhi Luo
Yuanyan Xiong
Xuegong Zhang
Hao Zhu
Hongwu Du
Liping Zhou
Z H E N G M Jiang
Sheng Mao
Y Shi
Pi Liu
Zehua Zeng
Xu Meng
LH Hu
Chen Li
Xuejiao Ma
Lianchong Gao
Xuelei Wang
Yawen Zheng
Jieshen Xian
Ziheng Lin
Yifan Lu
Wenzhuo Tang
Qiangwei Peng
Yuqing Ma
Cencan Xing
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%