Genos: a human-centric genomic foundation model
Anqi Lin, Bin Xie, Ye Cheng, Cheng Wang, Duoyuan Chen, Ercheng Wang, Fan Lü, Gui-Rong Xue, Hai-Qiang Zhang, Jiajie Zhan, Jianfeng Zhang, Jiangshuan Pang, Jianqiang Liang, Jiawei Lin, Jiaxin Ma, Jie Hu, Jing Ma, Jinni Dong, J. J. Li, Junchen Liu, Junhong Chen, Junyou Li, Kai Ding, Kaiwen Deng, Chen Kui, Lihui Wang, Longqi Liu, Ling Guo, Liwen Xiong, Luhao Yang, Ming Cheng, Ning Chen, Renzhong Chen, Shanxin Sun, Shaoshuai Li, Shicheng Chen, Shiping Liu, Siwei Xie, Suyan Liu, Tao Zhou, Wangyang Tang, Weiqiang Zhang, Xianyue Jiang, Xin Qi, Xin Jin, Xinjiang Tan, Xinyue Hu, Xun Xu, Xuyang Feng, Yafei Lü, Yifan Gao, Yongjia Shang, Youzhe He, Yue Yuan, Yufan Wang, Yuqi Liu, Zhan Xiao, Zhangyuan Meng, Zhaorong Li, Zhe Zhao, Zheng Yang, Zilin Wang
Abstract
BACKGROUND: The rapid expansion of human genomic data demands foundation models that manage ultra-long sequences and capture population diversity, limitations common in existing models that lack human-specific representation, and clinical inference efficiency. RESULTS: Here, we introduce Genos (Genos-1.2B/Genos-10B), a human-centric genomic foundation model engineered for million-basepair sequence modeling. Genos utilizes a large-scale mixture of experts structure, optimized for a 1-Mb context, trained on high-quality human de novo assemblies from datasets such as the Human Pangenome Reference Consortium and the Human Genome Structural Variation Consortium, representing diverse global populations. A suite of optimization strategies was implemented to ensure training stability and enhance computational efficiency, which collectively reduces costs and facilitates million-basepair context modeling. Functionally, Genos performs single-nucleotide resolution analysis and dynamically simulates the cascade effects of noncoding variations on RNA expression profiles. In comprehensive evaluations, Genos uniformly surpasses state-of-the-art models on critical human genomics benchmarks and demonstrates robust omics-text cross-modal diagnostic capabilities. We present a systematic technical evaluation and validation of Genos's architecture, training convergence, and performance across standard benchmarks. CONCLUSIONS: This work provides a reliable technical blueprint and performance benchmark for the development of the next generation of high-efficiency genomic foundation models. Genos model weights, inference code, and usage documentation are publicly available on GitHub (https://github.com/BGI-HangzhouAI/Genos) and Hugging Face Hub (https://huggingface.co/BGI-HangzhouAI).