Zhihao Zhan

GeneZip: Region-Aware Compression for Long Context DNA Modeling

Hongyu Guo

Genomic sequences span billions of base pairs (bp), posing a fundamental challenge for genome-scale foundation models. Existing approaches l… (see more)argely sidestep this barrier by either scaling relatively small models to long contexts or relying on heavy multi-GPU parallelism. Here we introduce GeneZip, a DNA compression model that leverages a key biological prior: genomic information is highly imbalanced. Coding regions comprise only a small fraction (about 2 percent) yet are information-dense, whereas most non-coding sequence is comparatively information-sparse. GeneZip couples HNet-style dynamic routing with a region-aware compression-ratio objective, enabling adaptive allocation of representation budget across genomic regions. As a result, GeneZip learns region-aware compression and achieves 137.6x compression with only 0.31 perplexity increase. On downstream long-context benchmarks, GeneZip achieves comparable or better performance on contact map prediction, expression quantitative trait loci prediction, and enhancer-target gene prediction. By reducing effective sequence length, GeneZip unlocks simultaneous scaling of context and capacity: compared to the prior state-of-the-art model JanusDNA, it enables training models 82.6x larger at 1M-bp context, supporting a 636M-parameter GeneZip model at 1M-bp context. All experiments in this paper can be trained on a single A100 80GB GPU.

2026-02-18

arXiv (preprint)

doi.org

openreview.net

GENERator: A Long-Context Generative Genomic Foundation Model

Q. Li

Wei Wu

Yong Zhang

Zhihao Zhan

Rui Chen

Mingyang Li

Kun Fu

Junyan Qi

Yongzhou Bao

Chao Wang

Yiheng Zhu

Zhiyun Zhang

Jian Tang

Fuli Feng

Jieping Ye

Liu Yuwen

Hui Xiong

Zheng Wang

Zhang, Yuanyuan

Chen, Ruipu … (see 2 more)

Wang, Chao

Tang, Jian

2026-02-03

Research Square (accepted)

doi.org

arxiv.org

Fast Proteome-Scale Protein Interaction Retrieval via Residue-Level Factorization

Narendra Chaudhary

Qian Cong

Jian Zhou

Sanchit Misra

Jian Tang

Protein-protein interactions (PPIs) are mediated at the residue level. Most sequence-based PPI models consider residue-residue interactions … (see more)across two proteins, which can yield accurate interaction scores but are too slow to scale. At proteome scale, identifying candidate PPIs requires evaluating nearly *all possible protein pairs*. For

2025-12-31

International Conference on Learning Representations (Accept (Poster))

openreview.net

Controllable Generation of Drug-like Molecules with Multi-modal Variational Flow

Fang Sun

Zhihao Zhan

Hongyu Guo

Ming Zhang

Jian Tang

Yizhou Sun

Designing drug molecules that bind effectively to target proteins while maintaining desired pharmacological properties remains a fundamental… (see more) challenge in drug discovery. Current approaches struggle to simultaneously control molecular topology and 3D geometry, often requiring expensive retraining for new design objectives. We propose a multi-modal variational flow framework that addresses these limitations by integrating a 2D topology encoder with a 3D geometry generator. Our architecture encodes molecular graphs into a learned latent distribution via junction tree representations, then employs normalizing flows to autoregressively generate atoms in 3D space conditioned on the protein binding site. This design enables zero-shot controllability: by manipulating the latent prior distribution, we can generate molecules with specific substructures or optimized properties without model retraining. Experiments on the CrossDocked benchmark show that our model achieves 31.1% high-affinity rate, substantially outperforming existing methods, while maintaining superior drug-likeness and structural diversity. Our framework opens new possibilities for on-demand molecular design, allowing medicinal chemists to rapidly explore chemical space with precise control over both structural motifs and physicochemical properties.

2025-10-21

logconference.io/LOG/2025/Conference (poster)

openreview.net

Overcoming Long-Context Limitations of State-Space Models via Context-Dependent Sparse Attention

Efficient long-context modeling remains a critical challenge for natural language processing (NLP), as the time complexity of the predominan… (see more)t Transformer architecture scales quadratically with the sequence length. While state-space models (SSMs) offer alternative sub-quadratic solutions, they struggle to capture long-range dependencies effectively. In this work, we focus on analyzing and improving the long-context modeling capabilities of SSMs. We show that the widely used synthetic task, associative recall, which requires a model to recall a value associated with a single key without context, insufficiently represents the complexities of real-world long-context modeling. To address this limitation, we extend the associative recall to a novel synthetic task, \emph{joint recall}, which requires a model to recall the value associated with a key given in a specified context. Theoretically, we prove that SSMs do not have the expressiveness to solve multi-query joint recall in sub-quadratic time complexity. To resolve this issue, we propose a solution based on integrating SSMs with Context-Dependent Sparse Attention (CDSA), which has the expressiveness to solve multi-query joint recall with sub-quadratic computation. To bridge the gap between theoretical analysis and real-world applications, we propose locality-sensitive Hashing Attention with sparse Key Selection (HAX), which instantiates the theoretical solution and is further tailored to natural language domains. Extensive experiments on both synthetic and real-world long-context benchmarks show that HAX consistently outperforms SSM baselines and SSMs integrated with context-independent sparse attention (CISA).

2025-09-17

NeurIPS.cc/2025/Conference (poster)

doi.org

openreview.net

Cell ontology guided transcriptome foundation model

Manqi Zhou

Boyu Han

Transcriptome foundation models (TFMs) hold great promises of deciphering the transcriptomic language that dictate diverse cell functions by… (see more) self-supervised learning on large-scale single-cell gene expression data, and ultimately unraveling the complex mechanisms of human diseases. However, current TFMs treat cells as independent samples and ignore the taxonomic relationships between cell types, which are available in cell ontology graphs. We argue that effectively leveraging this ontology information during the TFM pre-training can improve learning biologically meaningful gene co-expression patterns while preserving TFM as a general purpose foundation model for downstream zero-shot and fine-tuning tasks. To this end, we present **s**ingle **c**ell, **Cell**-**o**ntology guided TFM (scCello). We introduce cell-type coherence loss and ontology alignment loss, which are minimized along with the masked gene expression prediction loss during the pre-training. The novel loss component guide scCello to learn the cell-type-specific representation and the structural relation between cell types from the cell ontology graph, respectively. We pre-trained scCello on 22 million cells from CellxGene database leveraging their cell-type labels mapped to the cell ontology graph from Open Biological and Biomedical Ontology Foundry. Our TFM demonstrates competitive generalization and transferability performance over the existing TFMs on biologically important tasks including identifying novel cell types of unseen cells, prediction of cell-type-specific marker genes, and cancer drug responses. Source code and model weights are available at https://github.com/DeepGraphLearning/scCello.

2024-12-10

Neural Information Processing Systems (Accept (spotlight))

doi.org

openreview.net

Mila Ventures Launchpad

AI Policy Compass

Virtual Information Sessions

Zhihao Zhan

Publications

Mila Ventures Launchpad

AI Policy Compass

Virtual Information Sessions

Popular keywords:

Zhihao Zhan

Publications