Machine learning combining multi-omics data and network algorithms identifies adrenocortical carcinoma prognostic biomarkers - research

Aug 5, 2026

Upload by Finn

Adrenocortical carcinoma is a rare, aggressive endocrine cancer with an incomplete understanding of pathogenesis and limited therapeutic options. This study integrates machine learning with multi-omics and system-biology tools to discover prognostic biomarkers. Gene expression and DNA methylation datasets were analyzed using DIABLO for multi-omics signature discovery and latent-component integration, while network regulators were inferred via Clarivate CBDD network propagation and hidden-node methods. A random-forest model and Kaplan-Meier validation assessed discriminative and survival associations, producing a high-risk stratification signature with clinical relevance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Background

Methods

Data integration and signature discovery

Network propagation and regulator identification

Model training and prognostic evaluation

Results

Conclusion

Keywords

Introduction

TYPE Original Research PUBLISHED 06 November 2023 DOI 10.3389/fmolb.2023.1258902 OPEN ACCESS EDITED BY Silvia Bottini, Université Côte d’Azur, France REVIEWED BY Yize Li, Washington University in St. Louis, United States Tamás Micsik, Semmelweis University, Hungary *CORRESPONDENCE Roberto Martin-Hernandez, [roberto.martin@clarivate.com](roberto.martin@clarivate.com) †These authors have contributed equally to this work RECEIVED 14 July 2023 ACCEPTED 06 October 2023 PUBLISHED 06 November 2023 CITATION Martin-Hernandez R, Espeso-Gil S, Domingo C, Latorre P, Hervas S, Hernandez Mora JR and Kotelnikova E (2023), Machine learning combining multi-omics data and network algorithmsidentiﬁes adrenocortical carcinoma prognostic biomarkers. Front. Mol. Biosci. 10:1258902 . doi: 10.3389/fmolb.2023.1258902 COPYRIGHT © 2023 Martin-Hernandez, Espeso-Gil, Domingo, Latorre, Hervas, Hernandez Mora and Kotelnikova. This is an openaccess article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning combining multi-omics data and network algorithms identiﬁes adrenocortical carcinoma prognostic biomarkers Roberto Martin-Hernandez*, Sergio Espeso-Gil†, Clara Domingo†, Pablo Latorre, Sergi Hervas, Jose Ramon Hernandez Mora and Ekaterina Kotelnikova Discovery and Translational Sciences (DTS), Clarivate Analytics, Barcelona, Spain Background: Rare endocrine cancers such as Adrenocortical Carcinoma (ACC) present a serious diagnostic and prognostication challenge. The knowledge about ACC pathogenesis is incomplete, and patients have limited therapeutic options. Identiﬁcation of molecular drivers and effective biomarkers is required for timely diagnosis of the disease and stratify patients to offer the most beneﬁcial treatments. In this study we demonstrate how machine learning methods integrating multi-omics data, in combination with system biology tools, can contribute to the identiﬁcation of new prognostic biomarkers for ACC. Methods: ACC gene expression and DNA methylation datasets were downloaded from the Xena Browser (GDC TCGA Adrenocortical Carcinoma cohort) . A highly correlated multi-omics signature discriminating groups of samples was identiﬁed with the data integration analysis for biomarker discovery using latent components (DIABLO) method. Additional regulators of the identiﬁed signature were discovered using Clarivate CBDD (Computational Biology for Drug Discovery) network propagation and hidden nodes algorithms on a curated network of molecular interactions (MetaBase™) . The discriminative power of the multi-omics signature and their regulators was delineated by training a random forest classiﬁer using 55 samples, by employing a 10-fold cross validation with ﬁve iterations. The prognostic value of the identiﬁed biomarkers was further assessed on an external ACC dataset obtained from GEO (GSE49280) using the Kaplan-Meier estimator method. An optimal prognostic signature was ﬁnally derived using the stepwise Akaike Information Criterion (AIC) that allowed categorization of samples into high and low-risk groups. Results: A multi-omics signature including genes, micro RNA ’s and methylation sites was generated. Systems biology tools identiﬁed additional genes regulating the features included in the multi-omics signature. RNA-seq, miRNA-seq and DNA methylation sets of features revealed a high power to classify patients from stages I-II and stages III-IV, outperforming previously identiﬁed prognostic biomarkers. Using an independent dataset, associations of the genes included in the signature with Overall Survival (O

FAQs

What problem does the study address in adrenocortical carcinoma?

It targets the challenge of incomplete knowledge of ACC pathogenesis and the need for robust prognostic biomarkers to support timely diagnosis and patient stratification for more beneficial treatments.

How were the multi-omics data used to discover biomarkers?

The study downloaded ACC gene expression and DNA methylation datasets and used the DIABLO method to integrate multi-omics signals and identify a highly correlated signature discriminating sample groups.

How was prognostic value validated?

A random forest with 10-fold cross validation assessed discriminative power, and an external GEO dataset (GSE49280) was evaluated using Kaplan-Meier analysis. An optimal risk signature was then derived using stepwise AIC.

Aug 5, 2026

Upload by Finn