Gene-disease-drug link prediction using tripartite graphs
Cheng Chen, Stephen K. Grady, Sally R. Ellingson, Michael Allen Langston · 2021
The development of new ethical drugs is expensive in terms of both time and resources. A single drug can take up to a decade to bring to market, with costs soaring to over a billion dollars [1]. Drug repositioning has thus become an attractive alternative to the development of new compounds, with growing interest in the use of in silico repositioning predictions. Bipartite graphs and efficient biclique enumeration algorithms [2] can be used to study protein-drug or other crucial interactions. Extensions of this approach to larger dimensions has been hobbled, however, by a lack of effective analytics. In the present work, we take advantage of highly innovative and efficient tripartite graph algorithms [3]. We employ one partite set for genes, proteins or other gene products, another partite set for diseases, and a third partite set for drugs of interest, with inter-partite edges denoting known or inferred interactions.