Machine Learning and Omics Analysis in Aortic Aneurysm

Abstract

Aortic aneurysm is a life-threatening condition and mechanisms underlying its formation and progression are still incompletely understood. Omics approach has brought new insights to identify a broad spectrum of biomarkers and better understand cellular and molecular pathways involved. Omics generate a large amount of data and several studies have highlighted that artificial intelligence (AI) and techniques such as machine learning (ML)/deep learning (DL) can be of use in analyzing such complex datasets. However, only a few studies have so far reported the use of ML/DL for omics analysis in aortic aneurysms. The aim of this study is to summarize recent advances on the use of ML/DL for omics analysis to decipher aortic aneurysm pathophysiology and develop patient-tailored risk prediction models. In the light of current knowledge, we discuss current limits and highlight future directions in the field.

The key role of miRNA in syndromic and sporadic forms of ascending aortic aneurysms as biomarkers and targets of novel therapeutic strategies

“…Currently, the technological progresses and the reduction of their costs have allowed deep investigations of the pathogenetic mechanisms of many diseases, including aneurysms, for developing newest generation treatments ( Lareyre et al, 2023 ; Rega et al, 2023 ). In particular, the researchers are applying integrative multi-omics which appear to be necessary and show the potential to provide a greater quantity of information, ranging from the original causes of disease to the functional consequences or relevant interactions, and to clarify the physiological and pathophysiological role of miRNAs ( Lareyre et al, 2023 ; Rega et al, 2023 ).…”

“…Both analyses constitute a very serious challenge ( Recanatini and Menestrina, 2023 ), because firstly the interactions cannot be causal, since most links must be estimated through correlations or co-expressions; second, it is necessary to explore interactions between thousands or millions of entities (e.g., genes, epigenetic factors, proteins, metabolites, etc.) and this requires a huge amount of computational resources ( Lareyre et al, 2023 ; Rega et al, 2023 ). Regarding machine learning analysis, the first issue is data harmonization and dimensionality ( Lareyre et al, 2023 ; Rega et al, 2023 ).…”

“…and this requires a huge amount of computational resources ( Lareyre et al, 2023 ; Rega et al, 2023 ). Regarding machine learning analysis, the first issue is data harmonization and dimensionality ( Lareyre et al, 2023 ; Rega et al, 2023 ). Second, it is necessary to combine databases with different observations and characteristics; third, classification models need to be combined with a huge number of features in many samples ( Lareyre et al, 2023 ; Rega et al, 2023 ).…”

“…In this regard, epigenetic factors are promising candidates that may help overcome the problems mentioned above ( Kim and Stansfield, 2017 ; Knott and Doudna, 2018 ). Epigenetic mechanisms regulate gene expression without changing the DNA sequence ( Jiang et al, 2020 ; Wei et al, 2020 ; Lareyre et al, 2023 ). The latter include DNA methylation, post-translational histone modifications, and noncoding RNA ( Jiang et al, 2020 ; Wei et al, 2020 ; Han et al, 2023 ; Lareyre et al, 2023 ; Rega et al, 2023 ).…”

“…Epigenetic mechanisms regulate gene expression without changing the DNA sequence ( Jiang et al, 2020 ; Wei et al, 2020 ; Lareyre et al, 2023 ). The latter include DNA methylation, post-translational histone modifications, and noncoding RNA ( Jiang et al, 2020 ; Wei et al, 2020 ; Han et al, 2023 ; Lareyre et al, 2023 ; Rega et al, 2023 ). In recent years, many studies have investigated the role of epigenetic mechanisms in ATAA pathophysiology ( Chakraborty et al, 2023 ; Han et al, 2023 ; Rega et al, 2023 ; Tao et al, 2023 ).…”