1
|
Yu H, Zheng Y, Yang X. scDM: A deep generative method for cell surface protein prediction with diffusion model. J Mol Biol 2024; 436:168610. [PMID: 38754773 DOI: 10.1016/j.jmb.2024.168610] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/03/2024] [Revised: 05/06/2024] [Accepted: 05/09/2024] [Indexed: 05/18/2024]
Abstract
The executors of organismal functions are proteins, and the transition from RNA to protein is subject to post-transcriptional regulation; therefore, considering both RNA and surface protein expression simultaneously can provide additional evidence of biological processes. Cellular indexing of transcriptomes and epitopes by sequencing (CITE-Seq) technology can measure both RNA and protein expression in single cells, but these experiments are expensive and time-consuming. Due to the lack of computational tools for predicting surface proteins, we used datasets obtained with CITE-seq technology to design a deep generative prediction method based on diffusion models and to find biological discoveries through the prediction results. In our method, the scDM, which predicts protein expression values from RNA expression values of individual cells, uses a novel way of encoding the data into a model and generates predicted samples by introducing Gaussian noise to gradually remove the noise to learn the data distribution during the modelling process. Comprehensive evaluation across different datasets demonstrated that our predictions yielded satisfactory results and further demonstrated the effectiveness of incorporating information from single-cell multiomics data into diffusion models for biological studies. We also found that new directions for discovering therapeutic drug targets could be provided by jointly analysing the predictive value of surface protein expression and cancer cell drug scores.
Collapse
Affiliation(s)
- Hanlei Yu
- School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China
| | - Yuanjie Zheng
- School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China.
| | - Xinbo Yang
- School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China
| |
Collapse
|
2
|
Li Z, Liu G, Yang X, Shu M, Jin W, Tong Y, Liu X, Wang Y, Yuan J, Yang Y. An atlas of cell-type-specific interactome networks across 44 human tumor types. Genome Med 2024; 16:30. [PMID: 38347596 PMCID: PMC10860273 DOI: 10.1186/s13073-024-01303-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/05/2023] [Accepted: 02/06/2024] [Indexed: 02/15/2024] Open
Abstract
BACKGROUND Biological processes are controlled by groups of genes acting in concert. Investigating gene-gene interactions within different cell types can help researchers understand the regulatory mechanisms behind human complex diseases, such as tumors. METHODS We collected extensive single-cell RNA-seq data from tumors, involving 563 patients with 44 different tumor types. Through our analysis, we identified various cell types in tumors and created an atlas of different immune cell subsets across different tumor types. Using the SCINET method, we reconstructed interactome networks specific to different cell types. Diverse functional data was then integrated to gain biological insights into the networks, including somatic mutation patterns and gene functional annotation. Additionally, genes with prognostic relevance within the networks were also identified. We also examined cell-cell communications to investigate how gene interactions modulate cell-cell interactions. RESULTS We developed a data portal called CellNetdb for researchers to study cell-type-specific interactome networks. Our findings indicate that these networks can be used to identify genes with topological specificity in different cell types. We also found that prognostic genes can deconvolved into cell types through analyzing network connectivity. Additionally, we identified commonalities and differences in cell-type-specific networks across different tumor types. Our results suggest that these networks can be used to prioritize risk genes. CONCLUSIONS This study presented CellNetdb, a comprehensive repository featuring an atlas of cell-type-specific interactome networks across 44 human tumor types. The findings underscore the utility of these networks in delineating the intricacies of tumor microenvironments and advancing the understanding of molecular mechanisms underpinning human tumors.
Collapse
Affiliation(s)
- Zekun Li
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Gerui Liu
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Xiaoxiao Yang
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Meng Shu
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Wen Jin
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Yang Tong
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Xiaochuan Liu
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Yuting Wang
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China
| | - Jiapei Yuan
- State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, 300020, China.
- Tianjin Institutes of Health Science, Tianjin, 301600, China.
| | - Yang Yang
- Department of Bioinformatics, School of Basic Medical Sciences, The Province and Ministry Co-Sponsored Collaborative Innovation Center for Medical Epigenetics, Center for Reproductive Medicine, The Second Hospital of Tianjin Medical University, Tianjin Key Laboratory of Inflammatory Biology, Tianjin Medical University, Tianjin, 300070, China.
- Department of Pharmacology, School of Basic Medical Sciences, Tianjin Medical University, Tianjin, 300070, China.
| |
Collapse
|
3
|
Cha J, Lavi M, Kim J, Shomron N, Lee I. Imputation of single-cell transcriptome data enables the reconstruction of networks predictive of breast cancer metastasis. Comput Struct Biotechnol J 2023; 21:2296-2304. [PMID: 37035549 PMCID: PMC10073994 DOI: 10.1016/j.csbj.2023.03.036] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/14/2022] [Revised: 03/21/2023] [Accepted: 03/21/2023] [Indexed: 03/30/2023] Open
Abstract
Single-cell transcriptome data provide a unique opportunity to explore the gene networks of a particular cell type. However, insufficient capture rate and high dimensionality of single-cell RNA sequencing (scRNA-seq) data challenge cell-type-specific gene network (CGN) reconstruction. Here, we demonstrated that the imputation of scRNA-seq data enables reconstruction of CGNs by effective retrieval of gene functional associations. We reconstructed CGNs for seven primary and nine metastatic breast cancer cell lines using scRNA-seq data with imputation. Key genes for primary or metastatic cell lines were prioritized based on network centrality measures and CGN hub genes that were presumed to be the major determinant of cell type characteristics. To identify novel genes in breast cancer metastasis, we used the average rank difference of centrality between the primary and metastatic cell lines. Genes predicted using CGN centrality analysis were more enriched for known breast cancer metastatic genes than those predicted using differential expression. The molecular chaperone CCT2 was identified as a novel gene for breast metastasis during knockdown assays of several candidate genes. Overall, our study demonstrated an effective CGN reconstruction technique with imputation of scRNA-seq data and the feasibility of identifying key genes for particular cell subsets using single-cell network analysis.
Collapse
Affiliation(s)
- Junha Cha
- Department of Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
| | - Michael Lavi
- Faculty of Medicine and Edmond J Safra Center for Bioinformatics, Tel Aviv University, Tel Aviv 69978, Israel
| | - Junhan Kim
- Department of Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
| | - Noam Shomron
- Faculty of Medicine and Edmond J Safra Center for Bioinformatics, Tel Aviv University, Tel Aviv 69978, Israel
- Corresponding author.
| | - Insuk Lee
- Department of Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
- POSTECH Biotech Center, Pohang University of Science and Technology (POSTECH), Pohang 37673, Republic of Korea
- Corresponding author at: Department of Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea.
| |
Collapse
|