Urgent.News

What's breaking now, across thousands of outlets.

AI

A Biologically Informed Heterogeneous Graph Neural Network for Multi-Task Prediction of ncRNA-Metastasis-Cancer Interactions

Metastasis involves context-dependent molecular interactions in which non-coding RNAs, particularly miRNAs and circRNAs, play important regulatory roles. However, existing computational approaches generally do not jointly represent cancer type, metastatic event, and cancer-specific metastatic context. We developed a context-aware multi-task heterogeneous graph neural network (GNN) for predicting…

A novel computational approach has been developed to predict the associations between non-coding RNAs (ncRNAs) and cancer types, as well as metastatic events. This context-aware multi-task heterogeneous graph neural network (GNN) integrates various biological repositories to represent ncRNAs, cancers, metastatic event types (METs), and cancer-specific metastatic instances (CSMIs) in a single heterogeneous graph.

The model utilizes a hierarchical transformer-based encoder and a multi-relational TuckER decoder to perform six link-prediction tasks.

In independent runs, the model demonstrated an outstanding global AUROC of 0.8801 ± 0.0118 and an F1 score of 0.8260 ± 0.0071 when evaluated on a test set containing ncRNA-group-disjoint data. All three ablation variants, which trained the model on each task independently, resulted in lower AUROC scores, with the largest decrease observed under independent task training.

The framework was applied to pancreatic cancer, colorectal cancer, and hepatocellular carcinoma, providing disease-level, event-level, and expression-based evidence for the top-ranked candidate associations.

Overall, this context-aware heterogeneous graph neural network offers a powerful tool for prioritizing ncRNA-cancer-metastasis associations for experimental evaluation, ultimately aiding in the understanding and potential treatment of these complex diseases.

Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at biorxiv.org →

More in AI

More from Wednesday 19 August →