Finding melanoma drugs through a probabilistic knowledge graph

Abstract

Metastatic cutaneous melanoma is an aggressive skin cancer with some progressionslowing treatments but no known cure. The omics data explosion has created many possible drug candidates; however, filtering criteria remain challenging, and systems biology approaches have become fragmented with many disconnected databases. Using drug, protein and disease interactions, we built an evidence-weighted knowledge graph of integrated interactions. Our knowledge graph-based system, ReDrugS, can be used via an application programming interface or web interface, and has generated 25 high-quality melanoma drug candidates. We show that probabilistic analysis of systems biology graphs increases drug candidate quality compared to non-probabilistic methods. Four of the 25 candidates are novel therapies, three of which have been tested with other cancers. All other candidates have current or completed clinical trials, or have been studied in in vivo or in vitro. This approach can be used to identify candidate therapies for use in research or personalized medicine.

Enabling Web-scale data integration in biomedicine through Linked Open Data

“…Noor et al 49 constructed a mechanism-based DDI knowledge warehouse by integrating LSLOD content at the pharmacokinetic, pharmacodynamic, and pathway interaction level, and used an inference engine to generate mechanistic explanations for DDIs. ReDrugS 50 uses a data warehousing approach to integrate Bio2RDF Linked Data sources, and the integrated content is analyzed using a probabilistic graphical model to predict drug repurposing candidates for melanoma.…”

Quantitative comparison of the spreading and invasion of radial growth phase and metastatic melanoma cells in a three-dimensional human skin equivalent model

“…Melanoma is a deadly form of skin cancer ( Bertolotto, 2013 ; Weinstein et al, 2014 ) that is caused by the malignant transformation of melanocytes in the skin ( Uong & Zon, 2010 ; Bertolotto, 2013 ; Liu, Peng & Tobin, 2013 ). Melanoma accounts for less than 10% of all skin cancers, however it is associated with 80% of skin cancer related deaths ( Bandarchi et al, 2010 ; Bertolotto, 2013 ; Ramaraj & Cox, 2014 ; Leight et al, 2015 ; McCusker et al, 2017 ; Rivas et al, 2017 ). The early stage of a primary melanoma, where cancer cells are generally confined to the epidermis, is known as the radial growth phase (RGP) ( Clark, 1991 ; Meier et al, 2000 ).…”

A Specialized Recommender Agent for the Semantic Web

“…However, extending the representational power of OWL in such a way does not guarantee that reasoning with probabilistic OWL ontologies will be efficient or even decidable. A more feasible approach is taken in [10] where the underlying representation, the knowledge graph, provides contextual meaning to the search and is enhanced by the provenance of each fragment of knowledge captured used to compute their confidence probabilities. The logical framework based on context-dependent rules discussed in [11] has a similar motivation.…”