Generation and application of drug indication inference models using typed network motif comparison analysis

Abstract

BackgroundAs the amount of publicly available biomedical data increases, discovering hidden knowledge from biomedical data (i.e., Undiscovered Public Knowledge (UPK) proposed by Swanson) became an important research topic in the biological literature mining field. Drug indication inference, or drug repositioning, is one of famous UPK tasks, which infers alternative indications for approved drugs. Many previous studies tried to find novel candidate indications of existing drugs, but these works have following limitations: 1) models are not fully automated which required manual modulations to desired tasks, 2) are not able to cover various biomedical entities, and 3) have inference limitations that those works could infer only pre-defined cases using limited patterns. To overcome these problems, we suggest a new drug indication inference model.MethodsIn this paper, we adopted the Typed Network Motif Comparison Algorithm (TNMCA) to infer novel drug indications using topology of given network. Typed Network Motifs (TNM) are network motifs, which store types of data, instead of values of data. TNMCA is a powerful inference algorithm for multi-level biomedical interaction data as TNMs depend on the different types of entities and relations. We utilized a new normalized scoring function as well as network exclusion to improve the inference results. To validate our method, we applied TNMCA to a public database, Comparative Toxicogenomics Database (CTD).ResultsThe results show that enhanced TNMCA was able to infer meaningful indications with high performance (AUC = 0.801, 0.829) compared to the ABC model (AUC = 0.7050) and previous TNMCA model (AUC = 0.5679, 0.7469). The literature analysis also shows that TNMCA inferred meaningful results.ConclusionsWe proposed and enhanced a novel drug indication inference model by incorporating topological patterns of given network. By utilizing inference models from the topological patterns, we were able to improve inference power in drug indication inferences.

A CTD-Pfizer collaboration: manual curation of 88 000 scientific articles text mined for drug-disease and drug-phenotype interactions

“…In fact, curation from this project has already been incorporated into MetaADEDB, a new database of adverse drug events ( 30 , 31 ). As well, the dataset has been leveraged recently as a reference set to validate new algorithms for drug repositioning ( 32 ), as a standard for comparing successful drug–disease and drug–gene knowledge entity metrics ( 33 ), and as a resource for identifying chemical etiologies of diabetes ( 34 ). Additional improvements in text-mining strategies and manual biocuration will continue to enhance CTD as a premier resource for predictive toxicology.…”

Network analysis: a new approach to study endocrine disorders

“…Various mathematical operators are useful in pattern analysis. For example, network motif analysis identifies small networks that are over-represented when compared with a randomised version of the same network (Milo et al 2002), while network alignment and comparison are used to describe similarities between independent networks; this is a type of analysis that has been particularly valuable in understanding the evolutionarily conserved pathways that are common to several organisms (Choi et al 2013). Current software approaches to the assessment of network structure and its relationship to biological function are summarised in Table 1.…”

Network analysis and juvenile idiopathic arthritis (JIA): a new horizon for the understanding of disease pathogenesis and therapeutic target identification

“…“Network motif” analysis represents the identification of small networks, related to biological function, that are over-represented when compared with a randomized version of the same network [ 12 ]. Other important concepts in network analysis related to function include “node centrality” and “network robustness” [ 19 ] along with “network alignment and comparison”, an approach used to describe similarities between independent networks which has been particularly used to study the evolutionarily conserved pathways [ 20 ].…”