Deep BioNLP for Definition-based Embedding of Gene Products and Functional
Analysis of Genes
Many important applications in computational molecular biology such as
gene clustering, protein function prediction, protein interaction
evaluation and disease gene prioritization require functional similarity.
Moreover, to expedite the selection of candidate genes for gene-disease
research, genetic association studies, biomarker and drug target
selection, and animal models of human diseases, it is essential to have
search engines that can retrieve genes by their functions from proteome
databases. By the fast advancement in this domain, these engines can
substantially gain benefits from the functional similarity calculation of
gene products. In this talk, by restating the expressive power of deep
neural networks, we introduce a novel deep BioNLP model which efficiently
measures the functional similarity of proteins and gene products (e.g.
microRNA and mRNA). For this purpose, the model largely benefits from -
natural language definition of - (taxonomical) biological descriptors of
genes and the semantic similarity conveyed by these entities.