Shuofeng Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shuofeng Hu has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Xiaomin Ying
Center for Computational Biology, Beijing Institute of Basic Medical Sciences, Beijing, China
7 shared publications - Zhen He
Center for Computational Biology, Beijing Institute of Basic Medical Sciences, Beijing, China
5 shared publications Guohua Dong
Beijing Institute of Basic Medical Sciences
3 shared publications- Jiahao Zhou
Center for Computational Biology, Beijing Institute of Basic Medical Sciences, Beijing, China
3 shared publications - Jing Wang
Center for Computational Biology, Beijing Institute of Basic Medical Sciences, Beijing, China
3 shared publications - Jinhui Shi
Center for Computational Biology, Beijing Institute of Basic Medical Sciences, Beijing, China
3 shared publications
Latest Publications
- Open Access
- 23 September 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Neoadjuvant chemotherapy interferes with tertiary lymphoid structure formation in Epstein-Barr virus-associated gastric cancer
- 23 September 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessNeoadjuvant chemotherapy interferes with tertiary lymphoid structure formation in Epstein-Barr virus-associated gastric cancer
- Open Access
- 7 January 2026
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
Generalizable Single-cell Multimodal Data Integration with Self-supervised Learning
Recent breakthroughs in single-cell multi-omics technologies have enabled simultaneous measurement of diverse cellular modalities, offering unprecedented biological insights. However, integrating such multimodal data faces dual challenges: Small-scale paired-modality studies (hundreds of cells) risk overfitting, while large-scale reference atlases often struggle to generalize effectively to new datasets. To overcome these challenges, we present multimodal integration with self-supervised learning (MINERVA), a unified deep learning framework employing self-supervised strategies for single-cell multimodal integration. MINERVA outperforms six state-of-the-art methods in dimensionality reduction, missing feature imputation, and batch effect correction, even with limited training cells. For large-scale applications, MINERVA constructs scalable multi-tissue references that support zero-shot knowledge transfer to unseen datasets, instant cell type annotation, novel cell states identification, and comprehensive downstream analyses, all without model retraining. Uniquely bridging small-scale precision with atlas-level generalization, MINERVA serves as a versatile tool for both de novo integration and cost-effective atlas reuse in single-cell research.
- 7 January 2026
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
Open AccessGeneralizable Single-cell Multimodal Data Integration with Self-supervised Learning
Recent breakthroughs in single-cell multi-omics technologies have enabled simultaneous measurement of diverse cellular modalities, offering unprecedented biological insights. However, integrating such multimodal data faces dual challenges: Small-scale paired-modality studies (hundreds of cells) risk overfitting, while large-scale reference atlases often struggle to generalize effectively to new datasets. To overcome these challenges, we present multimodal integration with self-supervised learning (MINERVA), a unified deep learning framework employing self-supervised strategies for single-cell multimodal integration. MINERVA outperforms six state-of-the-art methods in dimensionality reduction, missing feature imputation, and batch effect correction, even with limited training cells. For large-scale applications, MINERVA constructs scalable multi-tissue references that support zero-shot knowledge transfer to unseen datasets, instant cell type annotation, novel cell states identification, and comprehensive downstream analyses, all without model retraining. Uniquely bridging small-scale precision with atlas-level generalization, MINERVA serves as a versatile tool for both de novo integration and cost-effective atlas reuse in single-cell research.
- Open Access
- 26 November 2025
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 24 (3)
DBP: Adaptive and Interpretable Factor Analysis for Single-cell RNA-seq Data with Deep Beta Processes
Factor analysis is a method that condenses multiple variables into a few latent factors. It can be used to extract the underlying sources of biological variation in high-dimensional data and distill them into interpretable gene programs. However, existing factorization methods lack adaptability in selecting the optimal number of factors and interpretability in capturing biological variation. To address these concerns, we propose Deep Beta Process (DBP), a deep probabilistic framework for adaptive and interpretable factor analysis of single-cell transcriptomic data. DBP achieves adaptive selection of factors through a stick-breaking Beta process and performs batch correction using an adversarial learning strategy. We validate the flexible factor extraction and robust batch correction capabilities of DBP on simulated datasets. We also demonstrate its superior performance in dimensionality reduction and biological interpretability while explaining biological variation from both cell and gene perspectives using factor and loading matrices. The application of DBP to a gastric adenocarcinoma dataset reveals malignant epithelial cell heterogeneity, providing valuable insights for investigating the molecular mechanisms of disease onset and progression. DBP is available at https://github.com/labomics/DBP and https://ngdc.cncb.ac.cn/biocode/tool/BT007954.
- 26 November 2025
- Published by Oxford University Press (OUP) in Genomics, Proteomics & Bioinformatics
- Vol. 24 (3)
Open AccessDBP: Adaptive and Interpretable Factor Analysis for Single-cell RNA-seq Data with Deep Beta Processes
Factor analysis is a method that condenses multiple variables into a few latent factors. It can be used to extract the underlying sources of biological variation in high-dimensional data and distill them into interpretable gene programs. However, existing factorization methods lack adaptability in selecting the optimal number of factors and interpretability in capturing biological variation. To address these concerns, we propose Deep Beta Process (DBP), a deep probabilistic framework for adaptive and interpretable factor analysis of single-cell transcriptomic data. DBP achieves adaptive selection of factors through a stick-breaking Beta process and performs batch correction using an adversarial learning strategy. We validate the flexible factor extraction and robust batch correction capabilities of DBP on simulated datasets. We also demonstrate its superior performance in dimensionality reduction and biological interpretability while explaining biological variation from both cell and gene perspectives using factor and loading matrices. The application of DBP to a gastric adenocarcinoma dataset reveals malignant epithelial cell heterogeneity, providing valuable insights for investigating the molecular mechanisms of disease onset and progression. DBP is available at https://github.com/labomics/DBP and https://ngdc.cncb.ac.cn/biocode/tool/BT007954.
- Open Access
Continual integration of single-cell multimodal data with MIRACLE
Single-cell sequencing technologies have revolutionized our understanding of cellular heterogeneity and facilitated the construction of multi-omics cell atlases via data integration. However, updating these atlases with new data conventionally requires reintegration of all data and is computationally intensive, hindering timely updates and dynamic adjustments in biological and medical research. To address this challenge, we present Multimodal Integration with Continual Learning (MIRACLE), a novel online learning framework for the adaptive and efficient integration of single-cell multimodal data. MIRACLE employs dynamic architectures and data rehearsal strategies to support continual learning, allowing diverse data to be integrated while minimizing information loss over time. Our evaluations demonstrate that MIRACLE achieves accurate online integration with reduced computational requirements, effectively updating and expanding atlases with new cross-tissue and cross-modal data, and precisely identifying novel cell types and transferring labels across datasets. MIRACLE provides an efficient and flexible tool for single-cell community to integrate, share and explore biological knowledge from single-cell multimodal data.
Open AccessContinual integration of single-cell multimodal data with MIRACLE
Single-cell sequencing technologies have revolutionized our understanding of cellular heterogeneity and facilitated the construction of multi-omics cell atlases via data integration. However, updating these atlases with new data conventionally requires reintegration of all data and is computationally intensive, hindering timely updates and dynamic adjustments in biological and medical research. To address this challenge, we present Multimodal Integration with Continual Learning (MIRACLE), a novel online learning framework for the adaptive and efficient integration of single-cell multimodal data. MIRACLE employs dynamic architectures and data rehearsal strategies to support continual learning, allowing diverse data to be integrated while minimizing information loss over time. Our evaluations demonstrate that MIRACLE achieves accurate online integration with reduced computational requirements, effectively updating and expanding atlases with new cross-tissue and cross-modal data, and precisely identifying novel cell types and transferring labels across datasets. MIRACLE provides an efficient and flexible tool for single-cell community to integrate, share and explore biological knowledge from single-cell multimodal data.
- Top Cited PapersOpen Access
- 23 January 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 42 (10) , 1594-1605
Mosaic integration and knowledge transfer of single-cell multimodal data with MIDAS
Integrating single-cell datasets produced by multiple omics technologies is essential for defining cellular heterogeneity. Mosaic integration, in which different datasets share only some of the measured modalities, poses major challenges, particularly regarding modality alignment and batch effect removal. Here, we present a deep probabilistic framework for the mosaic integration and knowledge transfer (MIDAS) of single-cell multimodal data. MIDAS simultaneously achieves dimensionality reduction, imputation and batch correction of mosaic data by using self-supervised modality alignment and information-theoretic latent disentanglement. We demonstrate its superiority to 19 other methods and reliability by evaluating its performance in trimodal and mosaic integration tasks. We also constructed a single-cell trimodal atlas of human peripheral blood mononuclear cells and tailored transfer learning and reciprocal reference mapping schemes to enable flexible and accurate knowledge transfer from the atlas to new data. Applications in mosaic integration, pseudotime analysis and cross-tissue knowledge transfer on bone marrow mosaic datasets demonstrate the versatility and superiority of MIDAS. MIDAS is available at https://github.com/labomics/midas.
- 23 January 2024
- Published by Springer Nature in Nature Biotechnology
- Vol. 42 (10) , 1594-1605
Top Cited PapersOpen AccessMosaic integration and knowledge transfer of single-cell multimodal data with MIDAS
Integrating single-cell datasets produced by multiple omics technologies is essential for defining cellular heterogeneity. Mosaic integration, in which different datasets share only some of the measured modalities, poses major challenges, particularly regarding modality alignment and batch effect removal. Here, we present a deep probabilistic framework for the mosaic integration and knowledge transfer (MIDAS) of single-cell multimodal data. MIDAS simultaneously achieves dimensionality reduction, imputation and batch correction of mosaic data by using self-supervised modality alignment and information-theoretic latent disentanglement. We demonstrate its superiority to 19 other methods and reliability by evaluating its performance in trimodal and mosaic integration tasks. We also constructed a single-cell trimodal atlas of human peripheral blood mononuclear cells and tailored transfer learning and reciprocal reference mapping schemes to enable flexible and accurate knowledge transfer from the atlas to new data. Applications in mosaic integration, pseudotime analysis and cross-tissue knowledge transfer on bone marrow mosaic datasets demonstrate the versatility and superiority of MIDAS. MIDAS is available at https://github.com/labomics/midas.
- Open Access
MIDAS: a deep generative model for mosaic integration and knowledge transfer of single-cell multimodal data
A bstract: Rapidly developing single-cell multi-omics sequencing technologies generate increasingly large bodies of multimodal data. Integrating multimodal data from different sequencing technologies, i.e . mosaic data, permits larger-scale investigation with more modalities and can help to better reveal cellular heterogeneity. However, mosaic integration involves major challenges, particularly regarding modality alignment and batch effect removal. Here we present a deep probabilistic framework for the mosaic integration and knowledge transfer (MIDAS) of single-cell multimodal data. MIDAS simultaneously achieves dimensionality reduction, imputation, and batch correction of mosaic data by employing self-supervised modality alignment and information-theoretic latent disentanglement. We demonstrate its superiority to other methods and reliability by evaluating its performance in full trimodal integration and various mosaic tasks. We also constructed a single-cell trimodal atlas of human peripheral blood mononuclear cells (PBMCs), and tailored transfer learning and reciprocal reference mapping schemes to enable flexible and accurate knowledge transfer from the atlas to new data. Applications in mosaic integration, pseudotime analysis, and cross-tissue knowledge transfer on bone marrow mosaic datasets demonstrate the versatility and superiority of MIDAS.
Open AccessMIDAS: a deep generative model for mosaic integration and knowledge transfer of single-cell multimodal data
A bstract: Rapidly developing single-cell multi-omics sequencing technologies generate increasingly large bodies of multimodal data. Integrating multimodal data from different sequencing technologies, i.e . mosaic data, permits larger-scale investigation with more modalities and can help to better reveal cellular heterogeneity. However, mosaic integration involves major challenges, particularly regarding modality alignment and batch effect removal. Here we present a deep probabilistic framework for the mosaic integration and knowledge transfer (MIDAS) of single-cell multimodal data. MIDAS simultaneously achieves dimensionality reduction, imputation, and batch correction of mosaic data by employing self-supervised modality alignment and information-theoretic latent disentanglement. We demonstrate its superiority to other methods and reliability by evaluating its performance in full trimodal integration and various mosaic tasks. We also constructed a single-cell trimodal atlas of human peripheral blood mononuclear cells (PBMCs), and tailored transfer learning and reciprocal reference mapping schemes to enable flexible and accurate knowledge transfer from the atlas to new data. Applications in mosaic integration, pseudotime analysis, and cross-tissue knowledge transfer on bone marrow mosaic datasets demonstrate the versatility and superiority of MIDAS.
- Open Access
MetaLogo: a heterogeneity-aware sequence logo generator and aligner
Sequence logos are used to visually display conservations and variations in short sequences. They can indicate the fixed patterns or conserved motifs in a batch of DNA or protein sequences. However, most of the popular sequence logo generators are based on the assumption that all the input sequences are from the same homologous group, which will lead to an overlook of the heterogeneity among the sequences during the sequence logo making process. Heterogeneous groups of sequences may represent clades of different evolutionary origins, or genes families with different functions. Therefore, it is essential to divide the sequences into different phylogenetic or functional groups to reveal their specific sequence motifs and conservation patterns. To solve these problems, we developed MetaLogo, which can automatically cluster the input sequences after multiple sequence alignment and phylogenetic tree construction, and then output sequence logos for multiple groups and aligned them in one figure. User-defined grouping is also supported by MetaLogo to allow users to investigate functional motifs in a more delicate and dynamic perspective. MetaLogo can highlight both the homologous and nonhomologous sites among sequences. MetaLogo can also be used to annotate the evolutionary positions and gene functions of unknown sequences, together with their local sequence characteristics. We provide users a public MetaLogo web server ( http://metalogo.omicsnet.org ), a standalone Python package ( https://github.com/labomics/MetaLogo ), and also a built-in web server available for local deployment. Using MetaLogo, users can draw informative, customized and publishable sequence logos without any programming experience to present and investigate new knowledge on specific sequence sets.
Open AccessMetaLogo: a heterogeneity-aware sequence logo generator and aligner
Sequence logos are used to visually display conservations and variations in short sequences. They can indicate the fixed patterns or conserved motifs in a batch of DNA or protein sequences. However, most of the popular sequence logo generators are based on the assumption that all the input sequences are from the same homologous group, which will lead to an overlook of the heterogeneity among the sequences during the sequence logo making process. Heterogeneous groups of sequences may represent clades of different evolutionary origins, or genes families with different functions. Therefore, it is essential to divide the sequences into different phylogenetic or functional groups to reveal their specific sequence motifs and conservation patterns. To solve these problems, we developed MetaLogo, which can automatically cluster the input sequences after multiple sequence alignment and phylogenetic tree construction, and then output sequence logos for multiple groups and aligned them in one figure. User-defined grouping is also supported by MetaLogo to allow users to investigate functional motifs in a more delicate and dynamic perspective. MetaLogo can highlight both the homologous and nonhomologous sites among sequences. MetaLogo can also be used to annotate the evolutionary positions and gene functions of unknown sequences, together with their local sequence characteristics. We provide users a public MetaLogo web server ( http://metalogo.omicsnet.org ), a standalone Python package ( https://github.com/labomics/MetaLogo ), and also a built-in web server available for local deployment. Using MetaLogo, users can draw informative, customized and publishable sequence logos without any programming experience to present and investigate new knowledge on specific sequence sets.