_version_ 1866912555325194240
author Zhang, Gongbo
Li, Yanting
Luo, Renqian
Hu, Pipi
Yang, Yang
Zhao, Zeru
Li, Lingbo
Liu, Guoqing
Wang, Zun
Bi, Ran
Gao, Kaiyuan
Guo, Liya
Xie, Yu
Liu, Chang
Zhang, Jia
Xie, Tian
Pinsler, Robert
Zeni, Claudio
Lu, Ziheng
Hao, Hongxia
Xia, Yingce
Segler, Marwin
Riechert, Maik
Yang, Wei
Jiang, Hao
Zhang, Wen-Bin
Zeng, Zhijun
Zhu, Yi
Dong, Li
Hu, Xiuyuan
Yuan, Li
Chen, Lei
Liu, Haiguang
Qin, Tao
author_facet Zhang, Gongbo
Li, Yanting
Luo, Renqian
Hu, Pipi
Yang, Yang
Zhao, Zeru
Li, Lingbo
Liu, Guoqing
Wang, Zun
Bi, Ran
Gao, Kaiyuan
Guo, Liya
Xie, Yu
Liu, Chang
Zhang, Jia
Xie, Tian
Pinsler, Robert
Zeni, Claudio
Lu, Ziheng
Hao, Hongxia
Xia, Yingce
Segler, Marwin
Riechert, Maik
Yang, Wei
Jiang, Hao
Zhang, Wen-Bin
Zeng, Zhijun
Zhu, Yi
Dong, Li
Hu, Xiuyuan
Yuan, Li
Chen, Lei
Liu, Haiguang
Qin, Tao
contents Function in natural systems arises from one-dimensional sequences forming three-dimensional structures with specific properties. However, current generative models suffer from critical limitations: training objectives seldom target function directly, discrete sequences and continuous coordinates are optimized in isolation, and conformational ensembles are under-modeled. We present UniGenX, a unified generative foundation model that addresses these gaps by co-generating sequences and coordinates under direct functional and property objectives across proteins, molecules, and materials. UniGenX represents heterogeneous inputs as a mixed stream of symbolic and numeric tokens, where a decoder-only autoregressive transformer provides global context and a conditional diffusion head generates numeric fields steered by task-specific tokens. Besides the new high SOTAs on structure prediction tasks, the model demonstrates state-of-the-art or competitive performance for the function-aware generation across domains: in materials, it achieves "conflicted" multi-property conditional generation, yielding 436 crystal candidates meeting triple constraints, including 11 with novel compositions; in chemistry, it sets new benchmarks on five property targets and conformer ensemble generation on GEOM; and in biology, it improves success in modeling protein induced fit (RMSD < 2 Å) by over 23-fold and enhances EC-conditioned enzyme design. Ablation studies and cross-domain transfer substantiate the benefits of joint discrete-continuous training, establishing UniGenX as a significant advance from prediction to controllable, function-aware generation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_06687
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UniGenX: a unified generative foundation model that couples sequence, structure and function to accelerate scientific design across proteins, molecules and materials
Zhang, Gongbo
Li, Yanting
Luo, Renqian
Hu, Pipi
Yang, Yang
Zhao, Zeru
Li, Lingbo
Liu, Guoqing
Wang, Zun
Bi, Ran
Gao, Kaiyuan
Guo, Liya
Xie, Yu
Liu, Chang
Zhang, Jia
Xie, Tian
Pinsler, Robert
Zeni, Claudio
Lu, Ziheng
Hao, Hongxia
Xia, Yingce
Segler, Marwin
Riechert, Maik
Yang, Wei
Jiang, Hao
Zhang, Wen-Bin
Zeng, Zhijun
Zhu, Yi
Dong, Li
Hu, Xiuyuan
Yuan, Li
Chen, Lei
Liu, Haiguang
Qin, Tao
Machine Learning
Materials Science
Artificial Intelligence
Biological Physics
Chemical Physics
Function in natural systems arises from one-dimensional sequences forming three-dimensional structures with specific properties. However, current generative models suffer from critical limitations: training objectives seldom target function directly, discrete sequences and continuous coordinates are optimized in isolation, and conformational ensembles are under-modeled. We present UniGenX, a unified generative foundation model that addresses these gaps by co-generating sequences and coordinates under direct functional and property objectives across proteins, molecules, and materials. UniGenX represents heterogeneous inputs as a mixed stream of symbolic and numeric tokens, where a decoder-only autoregressive transformer provides global context and a conditional diffusion head generates numeric fields steered by task-specific tokens. Besides the new high SOTAs on structure prediction tasks, the model demonstrates state-of-the-art or competitive performance for the function-aware generation across domains: in materials, it achieves "conflicted" multi-property conditional generation, yielding 436 crystal candidates meeting triple constraints, including 11 with novel compositions; in chemistry, it sets new benchmarks on five property targets and conformer ensemble generation on GEOM; and in biology, it improves success in modeling protein induced fit (RMSD < 2 Å) by over 23-fold and enhances EC-conditioned enzyme design. Ablation studies and cross-domain transfer substantiate the benefits of joint discrete-continuous training, establishing UniGenX as a significant advance from prediction to controllable, function-aware generation.
title UniGenX: a unified generative foundation model that couples sequence, structure and function to accelerate scientific design across proteins, molecules and materials
topic Machine Learning
Materials Science
Artificial Intelligence
Biological Physics
Chemical Physics
url https://arxiv.org/abs/2503.06687