Emergence of DSS efforts in genomics: Past contributions and challenges

摘要

Large amounts of data in biomedical research (from clinical data to gene expression data) are being generated. Use of these data sets and their associated knowledge are essential to understand the biological mechanisms behind diseases. While patients' clinical data from EHR can help researchers accurately and appropriately trace the performance of various kinds of medicines on the patients, the microarray data for the same pool of patients can contain valuable information for discovery of disease-associated gene expression patterns and can help classify the patients. However, research in the area of integrating genomic data with clinical data is still in its infancy and is riddled with many challenges. Even though data and knowledge sets are easily available from genome sequences and protein structural data of organisms, they usually are of many different varieties. Integrating them for a better understanding of biological functions at all levels is complicated. If we want to obtain the full benefit of functional genomics, we need to find a seamless way to integrate large amounts of patient datasets with genomic datasets in the field of biomedicine. Few papers in the decision support systems (DSS) literature provide an overview of Genomic Clinical Decision Support (GCDS) challenges that span data, knowledge, input/output, and architecture/implementation. This paper presents a unique effort dedicated to providing a comprehensive listing and a concise description of the DSS methodological challenges that arise from integrating complex and massive-scale genomic data with Clinical Decision Support (CDS) systems.