AI-driven drug discovery and repurposing using multi-omics for myocardial infarction and heart failure

Area of Science:

• Cardiovascular research
• Computational biology
• Drug discovery

Background:

• Cardiovascular diseases (CVDs) are a major global health burden, with drug discovery hindered by complex pathologies.
• Existing drug discovery methods struggle with the heterogeneity of CVDs.
• Artificial intelligence (AI), especially explainable AI (XAI), shows promise for analyzing multi-omics data in this field.

Purpose of the Study:

• To review the potential of integrating AI with multi-omics data for cardiovascular drug discovery.
• To explore AI's role in identifying novel therapeutic targets and repurposing drugs for myocardial infarction (MI) and heart failure (HF).
• To highlight current challenges and future directions for AI in CVD drug development.

Main Methods:

• Review of current literature on AI and XAI applications in cardiovascular research.
• Analysis of multi-omics datasets (genomic, transcriptomic, epigenomic, proteomic, metabolomic) using AI-driven methods.
• Discussion of emerging approaches combining AI with mechanistic models.

Main Results:

• AI and XAI can analyze complex multi-omics data to unravel CVD pathologies and identify therapeutic targets.
• AI can enhance the precision of biomarker discovery and drug response prediction.
• Integration of AI with multi-omics data shows potential for accelerating drug discovery for MI and HF.

Conclusions:

• AI and XAI hold significant promise for advancing cardiovascular drug discovery, despite current limitations.
• Addressing challenges like data heterogeneity and model interpretability is crucial for clinical translation.
• Future research should focus on combining AI with mechanistic models for improved biological relevance and therapeutic outcomes in CVDs.