Clinical and genomics data integration using meta-dimensional approach

Abstract

Clinical and genomics datasets contain humongous amount of information which are used in their respective environments independently to produce new science or better explain existing approaches. The interaction of data between these two domains is very limited and, hence, the information is disseminated. These disparate datasets need to be integrated to consolidate scattered pieces of information into a unified knowledge base to support new research challenges. However, there is no platform available that allows integration of clinical and genomics datasets into a consistent and coherent data source and produce analytics from it. We propose a data integration model here which will be capable of integrating clinical and genomics datasets using metadimensional approaches and machine learning methods. Bayesian Networks, which are based on meta-dimensional approach, will be used to design a probabilistic data model, and Neural Networks, which are based on machine learning, will be used for classification and pattern recognition from integrated data. This integration will help to coalesce the genetic background of clinical traits which will be immensely beneficial to derive new research insights for drug designing or precision medicine.