PhenoLinker: Phenotype-Gene Link Prediction and Explanation using Heterogeneous Graph Neural Networks

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.

association biology challenge convolutional neural network cs.lg gene graph graph neural networks graphs human information link prediction network networks neural network neural networks novel prediction q-bio.gn relationship