JLJie Liu

Jie Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Jie Liu has a total of 253 co-authors. Here are the top co-authors by number of shared publications:

  • Fan Feng

    Division of Diabetes, Endocrinology and Metabolism, Department of Medicine, Vanderbilt University Medical Center, Nashville, TN 37240, United States

    12 shared publications
  • Zhenhao Zhang

    Gilbert S. Omenn Department of Computational Medicine & Bioinformatics, University of Michigan

    8 shared publications
  • Xin Luo

    Gilbert S. Omenn Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, MI 48109, United States

    7 shared publications
  • Fan Feng

    Gilbert S. Omenn Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, USA

    6 shared publications
  • William Stafford Noble

    Department of Genome Sciences, University of Washington

    6 shared publications
  • Yuanhao Huang

    Department of Computational Medicine and Bioinformatics, University of Michigan , Ann Arbor , MI 48109

    6 shared publications

Latest Publications

  • Open Access

    Reference-Guided De Novo Retrieval of 3D Chromatin Organizational Features with Representation Learning

    by Sean Moran, Jie Liu

    Chromosome conformation maps reveal spatial genome organization through recurring contact patterns, some associated with established architectural features and others with less understood biological processes. Identifying instances of a pattern across these maps is necessary to investigate its biological associations and recurrence across experimental settings. However, conventional feature callers typically target predefined patterns, offering limited support when researchers can provide examples of a feature of interest but lack a dedicated detection method. We introduce Chromatin Fetch (C-Fetch), a reference-guided retrieval framework that uses examples of a desired feature and a contrasting class to rank and assign contact-map patches at provided candidate coordinates. C-Fetch learns a generalized representation through weakly KL-regularized 3D autoencoding of structurally related, non-temporal Hi-C and Micro-C patches without architectural feature labels. We demonstrate retrieval of biologically defined contacts, visually specified patterns, and established architectural features. Polycomb-guided retrieval identifies contacts enriched for PRC2-associated signals, exceeding enrichment from loop-guided assignments at 5 kb, while queries resembling condensed domains retrieve patches with progressively stronger rectangular enrichment at higher ranks. Loop-guided retrieval increases aggregate focal enrichment relative to complete Mustache proposal sets across four held-out maps, and selected contacts show greater focality loss after cohesin depletion than matched remaining candidates. Retrieval also extends to Region Capture Micro-C and HiChIP, with ranking agreement retained across biological replicates and reduced sequencing depths. Together, these results establish reference examples as a flexible means of specifying contact-map features of interest, enabling biological investigation, exploration of visual patterns, and refinement of existing calls across datasets and technologies.

  • Open Access

    A single-cell multimodal foundation model enables cross modality analysis between transcriptome and chromatin

    by Xinyu Bao, Jiajun Zhu, Xiangchen Song, Xiaoshan Zhang, Haoxuan Zeng, Ha T.H. Vu, Haoyun Zhang, Zheyu Zhang, Mingnan Yao, Mengdong Liang, et al.

    Most single-cell studies measure either gene expression or chromatin accessibility, and connecting the two usually requires fitting a model to each study. Here we present MUTUAL (MUltiome Transfer in a Unified Aligned Latent space), which keeps the single-cell foundation models scFoundation and EpiAgent frozen, combines their embeddings with linear summaries of each modality and learns a shared RNA– ATAC space from 3.0 million paired cells of 400 public multiome datasets. Applied without refitting to ten datasets not used for training, MUTUAL matched cells’ RNA and ATAC profiles more accurately than methods fitted on other donors of the same dataset, annotated ATAC-only cells more accurately and transferred transcription-factor motif activity to RNA-only cells beyond what cell-type assignment provides. In a maturation series of stem-cell-derived islets, motif activity predicted from RNA largely reproduced the measured stage-wise changes when other libraries of the series served as the reference. MUTUAL makes public multiome data directly usable by single-modality studies.

  • Open Access

    AutoPCR: automated phenotype concept recognition by prompting.

    by Yicheng Tao, Yuanhao Huang, Yiqun Wang, Xin Luo, Jie Liu

    Phenotype concept recognition (CR) is a fundamental task in biomedical text mining. However, existing methods either require ontology-specific training, making them struggle to generalize across diverse text styles and evolving biomedical terminology, or depend on general-purpose large language models that lack necessary domain knowledge. To address these limitations, we propose AutoPCR, a prompt-based phenotype CR method designed to automatically generalize to new ontologies and unseen data without ontology-specific training. To further boost performance, we also introduce an optional self-supervised training strategy. Experiments show that AutoPCR achieves the best average and most robust performance across datasets. Further ablation and transfer studies demonstrate its inductive capability and generalizability to new ontologies. Our code is available at https://github.com/yctao7/AutoPCR.

  • Open Access

    Deep-Plant: a supervised foundation model for plant regulatory genomics

    by Ahmed Daoud, Soumyadip Roy, Haoxuan Zeng, Xinyu Bao, Zhenhao Zhang, Jiakang Wang, Paul Parodi, Anireddy Sn Reddy, Jie Liu, Asa Ben-Hur

    Large-scale sequence-to-function deep learning models have demonstrated unparalleled ability to model biological sequences and have revolutionized the field of regulatory genomics. However, the majority of such efforts have centered on human and mammalian systems, leaving plant regulatory genomics comparatively underexplored. To address this gap, we introduce D eep -P lant , a supervised foundation model trained to predict chromatin state directly from genomic sequence. In contrast to large language models, which are trained in a selfsupervised manner using sequence alone, our model is trained to predict chromatin state across tissues and conditions. Training the model on a large collection of genome-wide experiments including DNA accessibility, transcription factor binding, and histone modifications, provides it with added biological context beyond the sequence itself. We demonstrate that the resulting model is an effective platform for developing accurate models of regulatory activity relevant to gene expression and active enhancers, exhibiting large improvements in speed, accuracy, and interpretability over the complementary approach of fine-tuning DNA language models. D eep -P lant models are available in Arabidopsis and rice, and work well as a building block for sequence modeling in related species such as corn. Together, these results establish supervised, chromatin-informed foundation models as a practical and effective paradigm for regulatory sequence modeling in plants.

  • Open Access

    The IGVF catalog—from genetic variation to function

    by Daofeng Li, Shane Liu, Pedro R Assis, Mingjie Li, Shengcheng Dong, Ian Whaling, Otto Jolanki, Meenakshi Kagda, Wenjin Zhang, Juan F Macias-Velasco, et al.

    Genomic variation between individuals is essential for understanding how differences in the genome sequence affect molecular and cellular processes. The Impact of Genomic Variation on Function (IGVF) Consortium aims to uncover the relationships among genomic variation, genome function, and phenotypes by combining experimental techniques, such as single-cell mapping and genomic perturbation assays, with computational approaches such as machine learning-based predictive modeling. The IGVF Data and Administrative Coordinating Centers collect, analyze, and disseminate data and results from across the consortium through an open-source platform called the IGVF Catalog. This resource includes, but is not limited to, data on the effects of coding variants on protein abundance and function, noncoding variants on enhancer activity (measured by MPRA or predicted computationally), and associations between variants and quantitative traits. All data are organized within a graph database comprising over 50 types of data collections with nearly 3 billion nodes and over 7.5 billion edges. The Catalog offers public API endpoints (https://api.catalogkg.igvf.org/) and a user-friendly interface for exploring, querying, and visualizing the data at https://catalog.igvf.org. We expect that this open-access platform will support the broader scientific community to advance our understanding of how genomic variation influences biology and disease.


  • Open Access

    CelLink: integrating single-cell multi-omics data with weak feature linkage and imbalanced cell populations.

    by Xin Luo, Yuanhao Huang, Haoxuan Zeng, Yicheng Tao, Xinyu Bao, Fan Feng, Alexander L Hopkirk, Thai Pham, Thomas S R Bate, Diane C Saunders, et al.

    Single-cell multi-omics technologies capture complementary molecular layers, enabling a comprehensive view of cellular states and functions. However, integrating these data types poses significant challenges when their features are weakly linked and cell population sizes are imbalanced. Currently, no method efficiently addresses these two issues simultaneously. Therefore, we developed CelLink, a novel single-cell multi-omics data integration method designed to overcome these challenges. CelLink normalizes and smooths feature profiles to align scales across datasets and integrates them through a multi-phase pipeline that iteratively employs the optimal transport algorithm. It dynamically refines cell-cell correspondences, identifying and excluding cells that cannot be reliably matched, thus avoiding performance degradation caused by erroneous imputations. This approach effectively adapts to weak feature linkage and imbalanced cell populations between datasets. Benchmarking CelLink on scRNA-seq and spatial proteomics datasets, as well as paired CITE-seq data, demonstrates its superior performance across various evaluation metrics, including data mixing, cell manifold structure preservation, and feature imputation accuracy. Compared to state-of-the-art methods, CelLink significantly outperforms others in imbalanced cell populations while consistently achieving better performance for balanced datasets. Moreover, CelLink uniquely enables cell subtype annotation, correction of mislabeled cells, and spatial transcriptomic analyses by imputing transcriptomic profiles for spatial proteomics data. Its great ability to impute large-scale paired single-cell multi-omics data positions it pivotal for building single-cell multi-modal foundation models and spatial cellular biology.


  • Open Access

    Developing a general AI model for integrating diverse genomic modalities and comprehensive genomic knowledge

    by Zhenhao Zhang, Xinyu Bao, Linghua Jiang, Xin Luo, Zheyu Zhang, Jing Yin, Meiqi Zhao, Yichun Wang, Annelise Comai, Joerg Waldhaus, et al.

    Advances in next-generation sequencing technologies have vastly expanded the availability of diverse genomic, epigenomic, and transcriptomic data, presenting the opportunity to develop a general AI model that integrates comprehensive genomic knowledge into a unified model. Unlike previous predictive models, which are typically specialized to certain tasks, our general AI model unifies a wide range of genomic modalities, such as nascent RNA and ultra-high-resolution chromatin organization, within a multi-task architecture. Using ATAC-seq and DNA sequences as inputs, we incorporated diverse genomic modalities as output, and the model exhibits strong generalizability across different cell types and tissues in all tasks we trained. It accurately predicts gene-level transcription measured by various nascent RNA assays, and effectively captures enhancer-associated transcription. Additionally, it accurately captures the potential functions of non-coding genetic variants and regulatory elements. Additionally, we extended the model trained on human data to a mouse general model, achieving accurate predictions of genomic modalities, such as high-resolution chromatin contact maps with limited data availability, which are further validated using an established mouse inner-ear study. This comprehensive approach offers a powerful tool for understanding genome regulation in both human and mouse species.


  • Open Access

    Statistically rigorous and computationally efficient chromatin stripe detection with Quagga

    by Fan Feng, Sean Moran, Anders S. Hansen, Xiaotian Zhang, Jie Liu

    Chromatin stripes are architectural chromatin features in which a singular loop anchor interacts with a contiguous region of DNA so, at the bulk sequencing level, it appears as a long stripe on chromatin contact matrices. Stripes are thought to play an important role in gene regulation and have been implicated in regulating a cell's lineage determination. Therefore, integrated analysis of stripes with genomic and epigenomic features at a genome-wide scale shows vast potential in understanding the cooperation between regulatory elements in 3D nucleome. To this end, we present Quagga, a computational tool for detection and statistical verification of genomic architectural stripes from Hi-C or Micro-C chromatin contact maps, that relies on robust image processing techniques and rigorous statistical tests for enrichment. Quagga outperforms other stripe detection methods in accuracy and is highly versatile, working with Hi-C, Micro-C, and other chromatin conformation capture data. By reporting on all tools’ performance in classifying CTCF-cohesin anchored stripes, enhancer–promoter interacting stripes, and indeterminate stripes, we also demonstrate a thorough, integrated analysis to determine the output stripes’ quality. Our work provides a flexible and convenient tool to help scientists explore the relationships between chromatin architectural stripes and important biological questions.


  • Open Access

    Retinoic acid receptor assembly dynamics governs dual functions in cochlear organogenesis

    by Saikat Chakraborty, Shuze Wang, Jack Ruhala, Brett Mehling, Jie Liu, Joerg Waldhaus

    Retinoic acid (RA) is a morphogen that contributes to inner ear development. Gain and loss of function experiments have indicated retinoic acid’s critical role in cochlear hair cell development. However, the underlying molecular mechanisms are unclear. Here, we hypothesized that RA receptor alpha (RARA) has a dual role in cochlear organogenesis: First, during embryonic development, in the presence of RA, RARA functions as a transcriptional activator that induces prosensory gene expression in progenitor cells and supports differentiation of the organ of Corti; later during postnatal development, when RA is absent, the function of RARA switches, thereby repressing prosensory genes in postnatal hair cells and hindering trans-differentiation into supporting cells. This hypothesis was supported by demonstration that RARA forms a complex with either the coactivator NCOA1 or the corepressor NCOR1 depending on the developmental stage. In addition, modulation of RA levels was found to govern recruitment of the coactivator and corepressor to the RARA complex, and the expression of prosensory genes was validated to depend on RARA complex composition. Together, our results provide insights supporting the potential of harnessing RA signaling to induce prosensory progenitors in stem cell–based strategies for hearing loss.