Abstract Body (Do not enter title and authors here): Introduction:
Heart failure is a significant concern for public health, particularly among the elderly demographic. In this study, we aimed to investigate the molecular underpinnings of heart failure, focusing on ferroptosis-related genes.
Purpose:
Our objective was to analyze unicellular RNA sequencing data derived from patients with heart failure and healthy controls. We sought to identify common ferroptosis-related genes and explore their potential as therapeutic targets for heart failure treatment.
Methods:
We analyzed unicellular RNA sequencing data from 200,615 cells, encompassing samples from four heart failure patients and two healthy controls. Initially, we curated a set of 388 ferroptosis-related genes from the FerrDB Database. Subsequently, we identified commonly shared significant ferroptosis-related genes by intersecting these with genes differentially expressed in the cell clusters using bulk RNAseq data. We also utilized data from the DGIdb to explore potential therapeutic drugs targeting these identified genes. Molecular docking and artificial intelligence-based IC50 prediction were employed to assess the efficacy of these drugs in mitigating heart failure.
Results:
Among the identified ferroptosis-related genes, ALOX5 and PLIN2 emerged as common candidates, both enriched in monocytes and dendritic cells. Interaction analysis revealed 51 drug categories interacting with ALOX5, along with 10 corresponding transcription factors. These genes exhibited robust discrimination ability in independent datasets, with an AUC of 0.915. Compounds exhibiting the strongest binding affinity with the ALOX5 protein were evaluated, demonstrating an affinity of 10.4 kcal/mol. Furthermore, a compound IC50 prediction system based on 42 machine learning models was developed, showing promising correlation coefficients up to 0.723 (P<0.05).
Conclusions:
Our research indicates ferroptosis may contribute to heart failure progression via ALOX5 and PLIN2 gene dysregulation in monocytes and dendritic cells, highlighting their therapeutic potential. Our study also emphasizes the value of computational approaches in drug discovery and personalized heart failure treatments.