The association of a given human phenotype to a genetic variant remains a critical challenge for biology. We present a novel system called PhenoLinker capable of associating a score to a phenotype-gene relationship by using heterogeneous information networks and a convolutional neural network-based model for graphs, which can provide an explanation for the predictions. This system can aid in the discovery of new associations and in the understanding of the consequences of human genetic variation.
@article{arxiv.2402.01809,
title = {PhenoLinker: Phenotype-Gene Link Prediction and Explanation using Heterogeneous Graph Neural Networks},
author = {Jose L. Mellina Andreu and Luis Bernal and Antonio F. Skarmeta and Mina Ryten and Sara Álvarez and Alejandro Cisterna García and Juan A. Botía},
journal= {arXiv preprint arXiv:2402.01809},
year = {2025}
}