Created Hetnet Connectivity Search to provide rapid insights into how two biomedical entities are related. The method identifies paths that occur more frequently than would be expected by chance (based on node degree alone) between two nodes. The study proposes matrix-multiplication based implementations for path count metrics and investigates null distributions from permuted networks.

From Himmelstein et al (2023) GigaScience

Go to the project page

Used hetnet edge prediction to predict whether each of 1,538 compounds treats each of 136 diseases. Browse this study's 209,168 predictions and the evidence behind them to explore new uses for existing drugs.

From Himmelstein et al (2017) eLife

Go to the project page

Used hetnet edge prediction for disease-associated genes, based on a predecessor network to Hetionet. Browse this study's 55,4364 results to prioritize genome-wide association studies by narrowing down the search.

From Himmelstein et al (2015) PLOS Computational Biology

Go to the project page

Explores the probability of edge existence due to node degree and provides a baseline for network-based predictions.

From Zietz et al (2024) GigaScience