Research Paper
Classification patterns identification of immunogenic cell death-related genes in heart failure based on deep learning
- 1 Department of Cardiology, Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai 200233, People’s Republic of China
Received: June 21, 2023 Accepted: December 26, 2023
https://doi.org/10.18632/aging.205620How to Cite
Copyright: © 2024 Ma et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Abstract
Heart failure (HF) is a complex and prevalent disease, especially among the elderly population, characterized by symptoms like chest tightness, shortness of breath, and dyspnea. To address the need for improved classification and drug target identification in HF, we explored the potential role of Immunogenic Cell Death (ICD), a mode of cell death known for its significance in the tumor immune response but relatively uncharted in HF research. In recent years, deep learning models have exhibited remarkable performance in tasks such as classification, clustering, and regression. In this paper, we harnessed the power of deep learning by employing various encoder models to evaluate their effectiveness in clustering based on ICD-related genes. This novel approach allowed us to identify distinct subtypes within HF. Subsequently, we refined these subtypes by employing differentially expressed genes, leading to the discovery of significant variations in immune infiltration and functional enrichment across these subtypes. Moreover, we leveraged advanced machine learning techniques to identify diagnosis-related genes in HF. The AUC of the diagnostic model in the internal and external test sets could reach more than 0.99. These genes served as the foundation for constructing nomogram models and further exploration of their interactions with miRNA and transcription factors. In summary, our study uniquely combines the exploration of ICD in HF, the application of deep learning models, and the identification of diagnosis-related genes to provide a multifaceted understanding of HF subtypes and potential therapeutic targets.