Integrative Analysis of Multi-Omics and Genetic Approaches-A New Level in Atherosclerotic Cardiovascular Risk Prediction

Genetics and environmental and lifestyle factors deeply affect cardiovascular diseases, with atherosclerosis as the etiopathological factor (ACVD) and their early recognition can significantly contribute to an efficient prevention and treatment of the disease. Due to the vast number of these factors, only the novel omic approaches are surmised. In addition to genomics, which extended the effective therapeutic potential for complex and rarer diseases, the use of omics presents a step-forward that can be harnessed for more accurate ACVD prediction and risk assessment in larger populations. The analysis of these data by artificial intelligence (AI)/machine learning (ML) strategies makes is possible to decipher the large amount of data that derives from such techniques, in order to provide an unbiased assessment of pathophysiological correlations and to develop a better understanding of the molecular background of ACVD. The predictive models implementing data from these omics , are based on consolidated AI best practices for classical ML and deep learning paradigms that employ methods (e.g., Integrative Network Fusion method, using an AI/ML supervised strategy and cross-validation) to validate the reproducibility of the results. Here, we highlight the proposed integrated approach for the prediction and diagnosis of ACVD with the presentation of the key elements of a joint scientific project of the University of Milan and the Almazov National Medical Research Centre.

Automated summary

Genetics, environmental and lifestyle factors greatly impact cardiovascular diseases, with atherosclerosis as the etiopathological factor. Novel omic approaches can enhance accurate prediction and risk assessment of ACVD, with the use of genomics expanding therapeutic potential and omics presenting a forward step in this direction. AI/ML strategies are essential for analyzing data and developing predictive models for ACVD diagnosis and prediction.