Using web services for linking genomic data to medical information systems
Using web services for linking genomic data to medical information systems.
Maojo V, Crespo J, de la Calle G, Barreiro J, Garcia-Remesal M. Using web services for linking genomic data to medical information systems. Methods Inf Med. 2007;46(4):484-92.
Introduction
Completion of the Human Genome Project has set the groundwork for the realization of personalized (or genomic) medicine. Biomedical information systems that link genetic and clinical data need to be developed in order to accelerate the advancement of personalized medicine and biomedical research.
Question
Can Web services be used to link clinical and genomic information for use in personalized practice and biomedical research?
Objective
The establishment of collaborative links by remote research sites using common software tools during The Human Genome Project ushered in a new approach to data access. Although this architecture worked well for genomic research, clinical databases must maintain confidentiality and therefore this model is inadequate for linking clinical and genomic data sets. Major roadblocks to achieving the integration of clinical and genomic data include collecting heterogeneous data from remote sources, the lack of interoperability, standards for data exchange, data modeling, user training and the incorporation of decision support tools for professionals using unfamiliar type of data. The utility of Web-service technologies for accessing remote heterogeneous information systems are described.
Results
The use of XML in Web based services creates a platform independent architecture for data exchange. Genomic databases (such as AGMIAL, NemaFootPrinter, and Genome Atlas Database) contain annotated genomic data along with various Web services including storage, text mining, and/or analysis tools. The integration of Web services is critical and several projects are underway to integrate different Web services. The European Union’s has supported the creation of Taverna which integrates web services into workflows. BioMOBY is an open source system based upon a decentralized architecture which stores Web service information in centralized registry. Gbrowse Moby is a web portal utilizing semantic Web service browsing to create linear workflows. OntoFusion is a system that integrates private and public databases with vocabulary and terminology databases to deliver access to heterogeneous data which is described in detail. The second generation system, called OntoFusion Plus, has been redesigned to function via 8 Web services which are described and graphically depicted.
Conclusions
Pilots of Web service based clinico-genomic frameworks have begun including the Early detection of Hereditary Non-Polyposis Colon Cancer (HNPCC) and the Advancing Clinico-Genomics Trials on Cancer (ACGT). The HNPCC is a Danish registry which contains genomic and epidemiologic information utilizing XML messaging to describe pedigree information, mutations, single nucleotide polymorphisms, and surveillance data. The goal of the ACGT is to establish a biomedical Grid infrastructure that would be accessible throughout Europe. The ACGT integrated virtual repository contains medical records, images, and genetic patient data.
Commentary
The results of this analysis are most useful to bioinformatics researchers, information technologists, medical researchers and clinicians. As personalized medicine becomes a standard of care, clinicians will require clinical decision support systems to assist in the interpretation and utilization of the newly available genetic data. Researchers will also need additional training to interpret clinical data and guidance on how to handle this unique, sensitive, and confidential data. Information technologists will be key in creating this seamless flow of protected and valuable data.
D. Przychodzin
