Hypotheses generation as supervised link discovery with automated class labeling on large-scale biomedical concept networks

Abstract

Computational approaches to generate hypotheses from biomedical literature have been studied intensively in recent years. Nevertheless, it still remains a challenge to automatically discover novel, cross-silo biomedical hypotheses from large-scale literature repositories. In order to address this challenge, we first model a biomedical literature repository as a comprehensive network of biomedical concepts and formulate hypotheses generation as a process of link discovery on the concept network. We extract the relevant information from the biomedical literature corpus and generate a concept network and concept-author map on a cluster using Map-Reduce frame-work. We extract a set of heterogeneous features such as random walk based features, neighborhood features and common author features. The potential number of links to consider for the possibility of link discovery is large in our concept network and to address the scalability problem, the features from a concept network are extracted using a cluster with Map-Reduce framework. We further model link discovery as a classification problem carried out on a training data set automatically extracted from two network snapshots taken in two consecutive time duration. A set of heterogeneous features, which cover both topological and semantic features derived from the concept network, have been studied with respect to their impacts on the accuracy of the proposed supervised link discovery process. A case study of hypotheses generation based on the proposed method has been presented in the paper.

Literature-based automated discovery of tumor suppressor p53 phosphorylation and inhibition by NEK2

“…This process relies on curated dictionaries and rules-based approaches to identify and normalize important biological entities ( 10 ). A pivotal demonstration of hypothesis generation from the biomedical literature is computer-aided discovery by Swanson linking—that is, if A causes B and B causes C, then A might cause C ( 11 – 13 )—the original example being between fish oil and Raynaud’s disease patients ( 14 ). More broadly, mining the literature for proteins, diseases, drugs, and their relationships allows for network-based approaches to identify disease biomarkers ( 15 ), repurpose drugs ( 16 ), and suggest protein function ( 17 ).…”

Discovering Research Hypotheses in Social Science Using Knowledge Graph Embeddings

“…Biomedical literature was also used by [10] to develop a link discovery method based on classification, where concepts are learnt and used as a basis for hypothesis generation. An Inductive Matrix Completion method was presented by [12], where the discovered gene-disease associations where supported by different types of evidence learnt as latent factors.…”

Leveraging graph topology and semantic context for pharmacovigilance through twitter-streams

“…Obviously, the manner in which drugs and effects are discussed can provide valuable information about their relationship. The notion of enriching a relationship graph with semantic context has been successfully employed in the context of biomedical literature mining [ 28 , 29 ].…”