# RNACOREX - RNA coregulatory network explorer and classifier

## TL;DR

RNACOREX is a new tool that helps find miRNA-mRNA networks linked to diseases and uses them to classify patient data with high interpretability.

## Contribution

RNACOREX introduces a user-friendly Python package combining curated databases and expression data to infer reliable miRNA–mRNA networks for disease classification.

## Key findings

- RNACOREX identifies disease-associated miRNA–mRNA networks using conditional mutual information and CLG classifiers.
- The tool achieves competitive classification performance across 13 cancer types from The Cancer Genome Atlas.
- It highlights miRNA–mRNA interactions consistently associated with survival outcomes in different cancers.

## Abstract

Micro-RNAs (miRNA) and their relationship with messenger RNAs (mRNA) have been widely associated with disease development and progression. Post-transcriptional coregulatory networks are sets of miRNA-mRNA interactions that regulate specific genetic behaviors through their combined activity. However, identifying reliable sets of such interactions associated with specific diseases remains challenging, partly due to the high rate of false positives and the lack of user-friendly tools developed for this purpose. In this work, we introduce a new Python package called RNACOREX (RNA CORegulatory network EXplorer and classifier). RNACOREX is a new, easy-to-use tool that allows researchers to find disease associated post-transcriptional coregulatory networks and use them to classify new unseen observations of miRNA and mRNA quantifications. RNACOREX combines structural information from curated databases with expression data analysis, using conditional mutual information to infer reliable sets of miRNA–mRNA interactions. These sets are then used to build probabilistic models based on Conditional Linear Gaussian (CLG) classifiers, which allow both prediction on new samples and validation of the inferred networks.

To demonstrate its capabilities, we tested RNACOREX in 13 different databases from the The Cancer Genome Atlas Program, generating the associated post-transcriptional coregulatory networks and extracting classification performance metrics for each tumor type. Specifically, we used RNACOREX to classify patients according to their survival time in each cancer type, highlighting miRNA–mRNA interactions that consistently appeared across different cancer types. The results show that RNACOREX achieves competitive predictive performance compared to widely used classification algorithms, while offering the added benefit of interpretability through its graph-based modeling framework.

Cells regulate their behavior through complex molecular processes, many of which are controlled by small molecules called microRNAs (miRNAs). These miRNAs bind to messenger RNAs (mRNAs) and influence how genes are expressed. When this regulation is disrupted, it can lead to diseases such as cancer. Rather than acting alone, miRNAs and mRNAs often work together in groups, forming networks that jointly control gene activity. Understanding these networks can help us better detect, classify, and study diseases.

In this work, we introduce RNACOREX, a new computational tool that helps researchers discover these miRNA–mRNA networks associated with specific diseases. RNACOREX combines prior biological knowledge from trusted databases with real gene expression data to identify reliable interaction patterns. It then uses these patterns to build predictive models that can classify new patient data and highlight the most relevant molecular interactions for each disease.

We tested RNACOREX on 13 different cancer types using data from The Cancer Genome Atlas. The tool revealed disease-associated networks and achieved performance comparable to other popular machine learning models. While the current implementation does not yet provide definitive biological insights, it highlights molecular interactions most associated with each phenotype, pointing to a promising direction for future investigations connecting prediction with biological context.