1
|
Xiao D, Lin M, Liu C, Geddes TA, Burchfield J, Parker B, Humphrey SJ, Yang P. SnapKin: a snapshot deep learning ensemble for kinase-substrate prediction from phosphoproteomics data. NAR Genom Bioinform 2023; 5:lqad099. [PMID: 37954574 PMCID: PMC10632189 DOI: 10.1093/nargab/lqad099] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/29/2023] [Revised: 09/18/2023] [Accepted: 10/25/2023] [Indexed: 11/14/2023] Open
Abstract
A major challenge in mass spectrometry-based phosphoproteomics lies in identifying the substrates of kinases, as currently only a small fraction of substrates identified can be confidently linked with a known kinase. Machine learning techniques are promising approaches for leveraging large-scale phosphoproteomics data to computationally predict substrates of kinases. However, the small number of experimentally validated kinase substrates (true positive) and the high data noise in many phosphoproteomics datasets together limit their applicability and utility. Here, we aim to develop advanced kinase-substrate prediction methods to address these challenges. Using a collection of seven large phosphoproteomics datasets, and both traditional and deep learning models, we first demonstrate that a 'pseudo-positive' learning strategy for alleviating small sample size is effective at improving model predictive performance. We next show that a data resampling-based ensemble learning strategy is useful for improving model stability while further enhancing prediction. Lastly, we introduce an ensemble deep learning model ('SnapKin') by incorporating the above two learning strategies into a 'snapshot' ensemble learning algorithm. We propose SnapKin, an ensemble deep learning method, for predicting substrates of kinases from large-scale phosphoproteomics data. We demonstrate that SnapKin consistently outperforms existing methods in kinase-substrate prediction. SnapKin is freely available at https://github.com/PYangLab/SnapKin.
Collapse
Affiliation(s)
- Di Xiao
- Computational Systems Biology Group, Children’s Medical Research Institute, The University of Sydney, Westmead, NSW 2145, Australia
| | - Michael Lin
- School of Mathematics and Statistics, The University of Sydney, Sydney, NSW 2006, Australia
| | - Chunlei Liu
- Computational Systems Biology Group, Children’s Medical Research Institute, The University of Sydney, Westmead, NSW 2145, Australia
| | - Thomas A Geddes
- Computational Systems Biology Group, Children’s Medical Research Institute, The University of Sydney, Westmead, NSW 2145, Australia
- Charles Perkins Centre, The University of Sydney, Sydney, NSW 2006, Australia
- School of Environmental and Life Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| | - James G Burchfield
- Charles Perkins Centre, The University of Sydney, Sydney, NSW 2006, Australia
- School of Environmental and Life Sciences, The University of Sydney, Sydney, NSW 2006, Australia
| | - Benjamin L Parker
- Centre for Muscle Research, Department of Anatomy and Physiology, School of Biomedical Sciences, Melbourne, VIC 3010, Australia
| | - Sean J Humphrey
- Charles Perkins Centre, The University of Sydney, Sydney, NSW 2006, Australia
- School of Environmental and Life Sciences, The University of Sydney, Sydney, NSW 2006, Australia
- Murdoch Children’s Research Institute, The Royal Children’s Hospital, Melbourne, VIC, 3052, Australia
| | - Pengyi Yang
- Computational Systems Biology Group, Children’s Medical Research Institute, The University of Sydney, Westmead, NSW 2145, Australia
- School of Mathematics and Statistics, The University of Sydney, Sydney, NSW 2006, Australia
- Charles Perkins Centre, The University of Sydney, Sydney, NSW 2006, Australia
| |
Collapse
|
2
|
Zittlau K, Nashier P, Cavarischia-Rega C, Macek B, Spät P, Nalpas N. Recent progress in quantitative phosphoproteomics. Expert Rev Proteomics 2023; 20:469-482. [PMID: 38116637 DOI: 10.1080/14789450.2023.2295872] [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: 08/11/2023] [Accepted: 12/12/2023] [Indexed: 12/21/2023]
Abstract
INTRODUCTION Protein phosphorylation is a critical post-translational modification involved in the regulation of numerous cellular processes from signal transduction to modulation of enzyme activities. Knowledge of dynamic changes of phosphorylation levels during biological processes, under various treatments or between healthy and disease models is fundamental for understanding the role of each phosphorylation event. Thereby, LC-MS/MS based technologies in combination with quantitative proteomics strategies evolved as a powerful strategy to investigate the function of individual protein phosphorylation events. AREAS COVERED State-of-the-art labeling techniques including stable isotope and isobaric labeling provide precise and accurate quantification of phosphorylation events. Here, we review the strengths and limitations of recent quantification methods and provide examples based on current studies, how quantitative phosphoproteomics can be further optimized for enhanced analytic depth, dynamic range, site localization, and data integrity. Specifically, reducing the input material demands is key to a broader implementation of quantitative phosphoproteomics, not least for clinical samples. EXPERT OPINION Despite quantitative phosphoproteomics is one of the most thriving fields in the proteomics world, many challenges still have to be overcome to facilitate even deeper and more comprehensive analyses as required in the current research, especially at single cell levels and in clinical diagnostics.
Collapse
Affiliation(s)
- Katharina Zittlau
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| | - Payal Nashier
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| | - Claudia Cavarischia-Rega
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| | - Boris Macek
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| | - Philipp Spät
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| | - Nicolas Nalpas
- Quantitative Proteomics, Interfaculty Institute of Cell Biology, University of Tuebingen, Tuebingen , Germany
| |
Collapse
|
3
|
Kim WS, Daddam JR, Keng BH, Kim J, Kim J. Heat shock protein 27 regulates myogenic and self-renewal potential of bovine satellite cells under heat stress. J Anim Sci 2023; 101:skad303. [PMID: 37688555 PMCID: PMC10629447 DOI: 10.1093/jas/skad303] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2023] [Accepted: 09/07/2023] [Indexed: 09/11/2023] Open
Abstract
While satellite cells play a key role in the hypertrophy, repair, and regeneration of skeletal muscles, their response to heat exposure remains poorly understood, particularly in beef cattle. This study aimed to investigate the changes in the transcriptome, proteome, and proliferation capability of bovine satellite cells in response to different levels of heat stress (HS) and exposure times. Satellite cells were isolated from 3-mo-old Holstein bulls (body weight: 77.10 ± 2.02 kg) and subjected to incubation under various temperature conditions: 1) control (38 °C; CON), 2) moderate (39.5 °C; MHS), and extreme (41 °C; EHS) for different durations ranging from 0 to 48 h. Following 3 h of exposure to extreme heat (EHS), satellite cells exhibited significantly increased gene expression and protein abundance of heat shock proteins (HSPs; HSP70, HSP90, HSP20) and paired box gene 7 (Pax7; P < 0.05). HSP27 expression peaked at 3 h of EHS and remained elevated until 24 h of exposure (P < 0.05). In contrast, the expression of myogenic factor 5 (Myf5) and paired box gene 3 (Pax3) was decreased by EHS compared to the control at 3 h of exposure (P < 0.05). Notably, the introduction of HSP27 small interference RNA (siRNA) transfection restored Myf5 expression to control levels, suggesting an association between HSP27 and Myf5 in regulating the self-renewal properties of satellite cells upon heat exposure. Immunoprecipitation experiments further confirmed the direct binding of HSP27 to Myf5, supporting its role as a molecular chaperone for Myf5. Protein-protein docking algorithms predicted a high probability of HSP27-Myf5 interaction as well. These findings indicate that extreme heat exposure intrinsically promotes the accumulation of HSPs and modulates the early myogenic regulatory factors in satellite cells. Moreover, HSP27 acts as a molecular chaperone by binding to Myf5, thereby regulating the division or differentiation of satellite cells in response to HS. The results of this study provide a better understanding of muscle physiology in heat-stressed cells, while unraveling the intricate molecular mechanisms that underlie the HS response in satellite cells.
Collapse
Affiliation(s)
- Won Seob Kim
- Department of Animal Science, Michigan State University, East Lansing, MI 48824, USA
| | - Jayasimha R Daddam
- Department of Animal Science, Michigan State University, East Lansing, MI 48824, USA
| | - Boon Hong Keng
- Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, USA
| | - Jaehwan Kim
- Department of Animal Science, Michigan State University, East Lansing, MI 48824, USA
| | - Jongkyoo Kim
- Department of Animal Science, Michigan State University, East Lansing, MI 48824, USA
- Department of Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, USA
- Animal Science and Food Science and Human Nutrition, Michigan State University, East Lansing, MI 48824, USA
| |
Collapse
|
4
|
Kao YH, Chen YJ, Higa S, Chattipakorn N, Santulli G. Editorial: Transcription factors and arrhythmogenesis. Front Physiol 2023; 14:1169747. [PMID: 36926195 PMCID: PMC10011700 DOI: 10.3389/fphys.2023.1169747] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2023] [Accepted: 02/21/2023] [Indexed: 03/08/2023] Open
Affiliation(s)
- Yu-Hsun Kao
- Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.,Department of Medical Education and Research, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan
| | - Yi-Jen Chen
- Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan.,Department of Internal Medicine, Division of Cardiovascular Medicine, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan
| | - Satoshi Higa
- Cardiac Electrophysiology and Pacing Laboratory, Division of Cardiovascular Medicine, Makiminato Central Hospital, Urasoe, Japan
| | - Nipon Chattipakorn
- Cardiac Electrophysiology Research and Training Center, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand
| | - Gaetano Santulli
- Department of Medicine, Wilf Family Cardiovascular Research Institute, Fleischer Institute for Diabetes and Metabolism (FIDAM), Albert Einstein College of Medicine, New York, NY, United States.,Einstein-Mount Sinai Diabetes Research Center (ES-DRC), Department of Molecular Pharmacology, Einstein Institute for Aging Research, Institute for Neuroimmunology and Inflammation (INI), Albert Einstein College of Medicine, New York, NY, United States
| |
Collapse
|