Knowledge Representation of G-Protein-Coupled Receptor Signal Transduction Pathways

G-protein-coupled receptors (GPCRs) are the largest family of plasma membrane receptors, which can be activated by an external signal such as a ligand. Binding of GPCRs and ligands in the plasma membrane activates pathways that involve a sequence of events. Better understanding of GPCRs and the signal transduction pathways can help biologists to target new drugs or regulate many important cellular functions for diseases. In this paper, we introduce ontology-based GPCR signal transduction pathways, which are converted from manually collected pathways in PubMed papers. We applied network graph embedding and knowledge graph embedding algorithms on the pathway data to discover protein interactions in the GPCR signal transduction pathways. Experiments show that we could suggest missing or unknown pathways by analyzing the ontology-based GPCR signal transduction pathways. Moreover, we introduced ontology constraints on the GPCR pathway for predicting the missing interactions. The experimental results showed that using ontology constraints can boost the performance of knowledge graph embedding algorithms.

[1]  M. Suwa,et al.  SEVENS: a database for comprehensive GPCR genes obtained from genomes , 2018, Biophysics and physicobiology.

[2]  Steven Skiena,et al.  DeepWalk: online learning of social representations , 2014, KDD.

[3]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[4]  Amar Koleti,et al.  GPCR ontology: development and application of a G protein-coupled receptor pharmacology knowledge framework , 2013, Bioinform..

[5]  Christophe Dessimoz,et al.  The Gene Ontology Handbook , 2017, Methods in Molecular Biology.

[6]  Jason Weston,et al.  Translating Embeddings for Modeling Multi-relational Data , 2013, NIPS.

[7]  Lorenzo Rosasco,et al.  Holographic Embeddings of Knowledge Graphs , 2015, AAAI.

[8]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[9]  Reinhard Schneider,et al.  GPCRs, G-proteins, effectors and their interactions: human-gpDB, a database employing visualization tools and data integration techniques , 2010, Database J. Biol. Databases Curation.

[10]  Ryutaro Ichise,et al.  TorusE: Knowledge Graph Embedding on a Lie Group , 2017, AAAI.

[11]  Linyuan Lü,et al.  Predicting missing links via local information , 2009, 0901.0553.

[12]  Jure Leskovec,et al.  node2vec: Scalable Feature Learning for Networks , 2016, KDD.

[13]  Kei-Hoi Cheung,et al.  BioPAX – A community standard for pathway data sharing , 2010, Nature Biotechnology.

[14]  Linyuan Lu,et al.  Link Prediction in Complex Networks: A Survey , 2010, ArXiv.

[15]  Dean Allemang,et al.  Semantic Web for the Working Ontologist - Effective Modeling in RDFS and OWL, Second Edition , 2011 .

[16]  Melinda R. Dwinell,et al.  The pathway ontology – updates and applications , 2014, Journal of Biomedical Semantics.

[17]  Guillaume Bouchard,et al.  Knowledge Graph Completion via Complex Tensor Factorization , 2017, J. Mach. Learn. Res..

[18]  Ubbo Visser,et al.  BioAssay Ontology (BAO): a semantic description of bioassays and high-throughput screening results , 2011, BMC Bioinformatics.

[19]  Tsuyoshi Murata,et al.  A General View for Network Embedding as Matrix Factorization , 2019, WSDM.