Zeyu Luo
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
Zeyu Luo has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Yu-Juan Zhang
Chongqing Normal University
4 shared publications - Junhao Liu
College of Life Science, Chongqing Normal University, Chongqing 401331, P. R. China
3 shared publications - Yawen Sun
College of Life Science, Chongqing Normal University, Chongqing 401331, P. R. China
3 shared publications - Dong-Qing Wei
School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai, China
2 shared publications - Guohua Wang
College of Computer and Control Engineering, Northeast Forestry University, Harbin, China
2 shared publications - Muhammad Tahir Khan
International Center for Interdisciplinary Research in Sciences (ICIRS), The University of Lahore, 1 KM Defence Road, Lahore, 58810, Pakistan; State Key Laboratory of Respiratory Disease, Guangzhou Key Laboratory of Tuberculosis Research, Department of Clinical Laboratory, Guangzhou Chest Hospital, Institute of Tuberculosis, Guangzhou Medical University, Guangzhou, China
2 shared publications
Latest Publications
- Open Access
- 21 November 2025
- Published by Elsevier in New Microbes and New Infections
- Vol. 68
Genomic landscape of Mycobacterium tuberculosis: Identifying mutation hotspots and stable regions for implications for drug development
- 21 November 2025
- Published by Elsevier in New Microbes and New Infections
- Vol. 68
Open AccessGenomic landscape of Mycobacterium tuberculosis: Identifying mutation hotspots and stable regions for implications for drug development
- Open Access
CTDP: Identifying cell types associated with disease phenotypes using scRNA-seq data
Single-cell RNA sequencing enables transcriptome-wide analysis at single-cell resolution, offering unprecedented insights into cellular heterogeneity across biological conditions. However, accurately comparing transcriptomic distributions of cells from distinct biological states, such as healthy versus diseased individuals, remains challenging. To address this, we developed CTDP, a robust and interpretable computational framework that identifies disease phenotype-associated cell types of interest by integrating Lasso-regularized logistic regression with permutation testing. Through comprehensive evaluations on both simulated and real-world datasets, including melanoma immunotherapy, COVID-19 severity, and liver cirrhosis, CTDP consistently outperformed existing methods such as DA-seq, scDist, and PENCIL in both accuracy and robustness. In melanoma, CTDP uncovered immune-responsive clusters and revealed transcriptional regulators like PTPRC, CREM, and JUNB linked to immunotherapy efficacy. In COVID-19, it identified critical severity-associated cell types, such as B cells, NK cells, epithelial cells, and macrophages, which contribute to dysregulated immune responses and inflammation in severe cases. These results highlight CTDP’s power in uncovering disease-relevant cell populations and its potential to advance precision medicine through single-cell analysis.
Open AccessCTDP: Identifying cell types associated with disease phenotypes using scRNA-seq data
Single-cell RNA sequencing enables transcriptome-wide analysis at single-cell resolution, offering unprecedented insights into cellular heterogeneity across biological conditions. However, accurately comparing transcriptomic distributions of cells from distinct biological states, such as healthy versus diseased individuals, remains challenging. To address this, we developed CTDP, a robust and interpretable computational framework that identifies disease phenotype-associated cell types of interest by integrating Lasso-regularized logistic regression with permutation testing. Through comprehensive evaluations on both simulated and real-world datasets, including melanoma immunotherapy, COVID-19 severity, and liver cirrhosis, CTDP consistently outperformed existing methods such as DA-seq, scDist, and PENCIL in both accuracy and robustness. In melanoma, CTDP uncovered immune-responsive clusters and revealed transcriptional regulators like PTPRC, CREM, and JUNB linked to immunotherapy efficacy. In COVID-19, it identified critical severity-associated cell types, such as B cells, NK cells, epithelial cells, and macrophages, which contribute to dysregulated immune responses and inflammation in severe cases. These results highlight CTDP’s power in uncovering disease-relevant cell populations and its potential to advance precision medicine through single-cell analysis.
- Open Access
- 2 July 2025
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 26 (4)
ESM2_AMP: an interpretable framework for protein–protein interactions prediction and biological mechanism discovery
The prediction of binary protein–protein interactions (PPIs) is essential for protein engineering, but a major challenge in deep learning-based methods is the unknown decision-making process of the model. To address this challenge, we propose the ESM2_AMP framework, which utilizes the ESM2 protein language model for extracting segment features from actual amino acid sequences and integrates the Transformer model for feature fusion in binary PPIs prediction. Further, the two distinct models, ESM2_AMPS and ESM2_AMP_CSE are developed to systematically explore the contributions of segment features and combine with special tokens features in the decision-making process. The experimental results reveal that the model relying on segment features demonstrates strong correlations between segments with high attention weights and known functional regions of amino acid sequences. This insight suggests that attention to these segments helps capture biologically relevant functional and interaction-related information. By analyzing the coverage relationship between high-attention sequence fragments and functional regions, we validated the model’s ability to capture key segment features of PPIs and revealed the critical role of functional domains in PPIs. This finding not only enhances the interpretability methods for sequence-based prediction models but also provides biological evidence supporting the important regulatory role of functional sequences in protein–protein interactions. It offers cross-disciplinary insights for algorithm optimization and experimental validation research in the field of computational biology.
- 2 July 2025
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 26 (4)
Open AccessESM2_AMP: an interpretable framework for protein–protein interactions prediction and biological mechanism discovery
The prediction of binary protein–protein interactions (PPIs) is essential for protein engineering, but a major challenge in deep learning-based methods is the unknown decision-making process of the model. To address this challenge, we propose the ESM2_AMP framework, which utilizes the ESM2 protein language model for extracting segment features from actual amino acid sequences and integrates the Transformer model for feature fusion in binary PPIs prediction. Further, the two distinct models, ESM2_AMPS and ESM2_AMP_CSE are developed to systematically explore the contributions of segment features and combine with special tokens features in the decision-making process. The experimental results reveal that the model relying on segment features demonstrates strong correlations between segments with high attention weights and known functional regions of amino acid sequences. This insight suggests that attention to these segments helps capture biologically relevant functional and interaction-related information. By analyzing the coverage relationship between high-attention sequence fragments and functional regions, we validated the model’s ability to capture key segment features of PPIs and revealed the critical role of functional domains in PPIs. This finding not only enhances the interpretability methods for sequence-based prediction models but also provides biological evidence supporting the important regulatory role of functional sequences in protein–protein interactions. It offers cross-disciplinary insights for algorithm optimization and experimental validation research in the field of computational biology.
- Open Access
- 26 June 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (7)
Multistage attention-based extraction and fusion of protein sequence and structural features for protein function prediction
Motivation: Protein function prediction is important for drug development and disease treatment. Recently, deep learning methods have leveraged protein sequence and structural information, achieving remarkable progress in the field of protein function prediction. However, existing methods ignore the complex multimodal interaction information between sequence and structural features. Since protein sequence and structural information reveal the functional characteristics of proteins from different perspectives, it is challenging to effectively fuse the information from these two modalities to portray protein functions more comprehensively. In addition, current methods have difficulty in effectively capturing long-range dependencies and global contextual information in protein sequences during feature extraction, thus limiting the ability of the model to recognize critical functional residues. Results: In this study, we propose a novel framework termed Multi-stage Attention-based Extraction and Fusion model for GO prediction (MAEF-GO) based on a multistage attention mechanism to predict protein functions. MAEF-GO innovatively integrates the graph convolutional network and the graph attention network to extract protein structural features. To address the issue of modeling long-range dependencies within protein sequences, we introduce a frequency-domain attention mechanism capable of extracting global contextual relationships. Additionally, a cross-attention module is implemented to facilitate interactive fusion between protein sequence and structural modalities. Experimental evaluations demonstrate that MAEF-GO achieves superior performance compared to several state-of-the-art baseline models across standard benchmarks. Furthermore, analysis of the cross-attention weight distributions demonstrates MAEF-GO’s interpretability. It can effectively identify critical functional residues of proteins. Availability and implementation: The MAEF-GO source code can be found at https://github.com/nebstudio/MAEF-GO, an archived snapshot of the code used in this study is also available via Zenodo at https://doi.org/10.5281/zenodo.15422392.
- 26 June 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (7)
Open AccessMultistage attention-based extraction and fusion of protein sequence and structural features for protein function prediction
Motivation: Protein function prediction is important for drug development and disease treatment. Recently, deep learning methods have leveraged protein sequence and structural information, achieving remarkable progress in the field of protein function prediction. However, existing methods ignore the complex multimodal interaction information between sequence and structural features. Since protein sequence and structural information reveal the functional characteristics of proteins from different perspectives, it is challenging to effectively fuse the information from these two modalities to portray protein functions more comprehensively. In addition, current methods have difficulty in effectively capturing long-range dependencies and global contextual information in protein sequences during feature extraction, thus limiting the ability of the model to recognize critical functional residues. Results: In this study, we propose a novel framework termed Multi-stage Attention-based Extraction and Fusion model for GO prediction (MAEF-GO) based on a multistage attention mechanism to predict protein functions. MAEF-GO innovatively integrates the graph convolutional network and the graph attention network to extract protein structural features. To address the issue of modeling long-range dependencies within protein sequences, we introduce a frequency-domain attention mechanism capable of extracting global contextual relationships. Additionally, a cross-attention module is implemented to facilitate interactive fusion between protein sequence and structural modalities. Experimental evaluations demonstrate that MAEF-GO achieves superior performance compared to several state-of-the-art baseline models across standard benchmarks. Furthermore, analysis of the cross-attention weight distributions demonstrates MAEF-GO’s interpretability. It can effectively identify critical functional residues of proteins. Availability and implementation: The MAEF-GO source code can be found at https://github.com/nebstudio/MAEF-GO, an archived snapshot of the code used in this study is also available via Zenodo at https://doi.org/10.5281/zenodo.15422392.
- 2 May 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Computational Biophysics and Chemistry
- Vol. 25 (01) , 1-23
ProtLoc-Mex1: Interpretable Analysis of Amino Acids Sequence Chemical Feature and GO Annotations for Predicting Protein Subcellular Localization
Machine learning algorithms have revolutionized the study of protein subcellular localization; however, their black-box nature limits their interpretability. To address this, we built upon ProtLoc-Mex1, an automated pipeline incorporating interpretation techniques, by integrating chemical and GO annotation features to create a transparent random forest predictor. When applied to membrane protein type distinction and subcellular localization prediction, ProtLoc-Mex1 identified important features, explored the feature interaction effect and explored the functional feature semantic representation in language models, deepening our understanding of protein targeting mechanisms. We also created modules to facilitate their use for various prediction understanding in machine learning systems and provide a valuable resource for the scientific community. The code is available at ( https://github.com/yujuan-zhang/ProtLoc-mexl ).
- 2 May 2025
- Published by World Scientific Pub Co Pte Ltd in Journal of Computational Biophysics and Chemistry
- Vol. 25 (01) , 1-23
ProtLoc-Mex1: Interpretable Analysis of Amino Acids Sequence Chemical Feature and GO Annotations for Predicting Protein Subcellular Localization
Machine learning algorithms have revolutionized the study of protein subcellular localization; however, their black-box nature limits their interpretability. To address this, we built upon ProtLoc-Mex1, an automated pipeline incorporating interpretation techniques, by integrating chemical and GO annotation features to create a transparent random forest predictor. When applied to membrane protein type distinction and subcellular localization prediction, ProtLoc-Mex1 identified important features, explored the feature interaction effect and explored the functional feature semantic representation in language models, deepening our understanding of protein targeting mechanisms. We also created modules to facilitate their use for various prediction understanding in machine learning systems and provide a valuable resource for the scientific community. The code is available at ( https://github.com/yujuan-zhang/ProtLoc-mexl ).
- 24 May 2024
- Published by Elsevier in Desalination
- Vol. 586, 117780
Preparation of positively charged nanofiltration membranes: Manipulation of the positive charge
- 24 May 2024
- Published by Elsevier in Desalination
- Vol. 586, 117780
Preparation of positively charged nanofiltration membranes: Manipulation of the positive charge
- Open Access
- 22 January 2024
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (2)
Interpretable feature extraction and dimensionality reduction in ESM2 for protein localization prediction
As the application of large language models (LLMs) has broadened into the realm of biological predictions, leveraging their capacity for self-supervised learning to create feature representations of amino acid sequences, these models have set a new benchmark in tackling downstream challenges, such as subcellular localization. However, previous studies have primarily focused on either the structural design of models or differing strategies for fine-tuning, largely overlooking investigations into the nature of the features derived from LLMs. In this research, we propose different ESM2 representation extraction strategies, considering both the character type and position within the ESM2 input sequence. Using model dimensionality reduction, predictive analysis and interpretability techniques, we have illuminated potential associations between diverse feature types and specific subcellular localizations. Particularly, the prediction of Mitochondrion and Golgi apparatus prefer segments feature closer to the N-terminal, and phosphorylation site-based features could mirror phosphorylation properties. We also evaluate the prediction performance and interpretability robustness of Random Forest and Deep Neural Networks with varied feature inputs. This work offers novel insights into maximizing LLMs’ utility, understanding their mechanisms, and extracting biological domain knowledge. Furthermore, we have made the code, feature extraction API, and all relevant materials available at https://github.com/yujuan-zhang/feature-representation-for-LLMs.
- 22 January 2024
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 25 (2)
Open AccessInterpretable feature extraction and dimensionality reduction in ESM2 for protein localization prediction
As the application of large language models (LLMs) has broadened into the realm of biological predictions, leveraging their capacity for self-supervised learning to create feature representations of amino acid sequences, these models have set a new benchmark in tackling downstream challenges, such as subcellular localization. However, previous studies have primarily focused on either the structural design of models or differing strategies for fine-tuning, largely overlooking investigations into the nature of the features derived from LLMs. In this research, we propose different ESM2 representation extraction strategies, considering both the character type and position within the ESM2 input sequence. Using model dimensionality reduction, predictive analysis and interpretability techniques, we have illuminated potential associations between diverse feature types and specific subcellular localizations. Particularly, the prediction of Mitochondrion and Golgi apparatus prefer segments feature closer to the N-terminal, and phosphorylation site-based features could mirror phosphorylation properties. We also evaluate the prediction performance and interpretability robustness of Random Forest and Deep Neural Networks with varied feature inputs. This work offers novel insights into maximizing LLMs’ utility, understanding their mechanisms, and extracting biological domain knowledge. Furthermore, we have made the code, feature extraction API, and all relevant materials available at https://github.com/yujuan-zhang/feature-representation-for-LLMs.
- 22 July 2023
- Published by World Scientific Pub Co Pte Ltd in Journal of Computational Biophysics and Chemistry
- Vol. 22 (06) , 673-686
Exploring Key Proteins, Pathways and Oxygen Usage Bias of Proteins and Metabolites in Melanoma
Hypoxia plays a critical role in melanoma development, but the characteristics of elemental oxygen in proteins and adaptation to hypoxia microenvironments are still unidentified. This study aims to explore oxygen contents (OCs) and differentially expressed proteins (DEP). Protein expression data were retrieved from Human Protein Atlas. The DEP in melanoma samples were compared with normal skin cells. We identified 1,969 DEP, and none of the genes coding these proteins were present on chromosome Y. The average oxygen content (AOC) was 7.24% higher in highly expressed proteins than lowly expressed proteins in melanoma and normal skin cells. The AOC is 2.36% higher in the up regulated proteins (URPs) in melanoma. The essential amino acids in the proteins in melanoma cells contributed to increased OC. Functional dissections of the high OCs in URP displayed that some of these proteins are associated with cytoskeleton, cyclins and cell cycle proteins. The URP interactions were generated using a STRING database. Majority of these URPs are associated in expression, exhibiting sufficient interactions with each other. This study provides useful information regarding protein expression in melanoma cells and the molecular mechanism of melanoma using stoichiogenomics.
- 22 July 2023
- Published by World Scientific Pub Co Pte Ltd in Journal of Computational Biophysics and Chemistry
- Vol. 22 (06) , 673-686
Exploring Key Proteins, Pathways and Oxygen Usage Bias of Proteins and Metabolites in Melanoma
Hypoxia plays a critical role in melanoma development, but the characteristics of elemental oxygen in proteins and adaptation to hypoxia microenvironments are still unidentified. This study aims to explore oxygen contents (OCs) and differentially expressed proteins (DEP). Protein expression data were retrieved from Human Protein Atlas. The DEP in melanoma samples were compared with normal skin cells. We identified 1,969 DEP, and none of the genes coding these proteins were present on chromosome Y. The average oxygen content (AOC) was 7.24% higher in highly expressed proteins than lowly expressed proteins in melanoma and normal skin cells. The AOC is 2.36% higher in the up regulated proteins (URPs) in melanoma. The essential amino acids in the proteins in melanoma cells contributed to increased OC. Functional dissections of the high OCs in URP displayed that some of these proteins are associated with cytoskeleton, cyclins and cell cycle proteins. The URP interactions were generated using a STRING database. Majority of these URPs are associated in expression, exhibiting sufficient interactions with each other. This study provides useful information regarding protein expression in melanoma cells and the molecular mechanism of melanoma using stoichiogenomics.