SWC Banner
Home | Objectives | SW-Application | Participation | Challenge 2003 | Examples | Submissions 2003 | Sponsors | Contact |


Exploiting Bioinformatics Web Resources for Single Nucleotide Polymorphism (SNP) Analysis

Single nucleotide polymorphism (SNP) markers are single base pair positions in genomic DNA at which different sequence alternatives (alleles) exist in normal individuals in some population(s), wherein the least frequent allele has an abundance of one percent or greater. SNP alleles can be used as genetic markers. Because the SNP itself is the variant that causes or contributes to the risk of developing a particular genetic disorder, SNPs are expected to facilitate large-scale association genetics studies, which usually aim at identifying novel disease-causing genes and possible treatments of many genetic disorders. To perform association study requires integrating a variety of information sources. However, the number of biological on-line databases and tools is growing at breakneck speed. To integrate these databases and tools to provide a convenient search tool is a challenging task in bioinformatics. Thus, solutions for database integration and interoperability of bioinformatics tools are in urgent need. In this project, we apply the Semantic Web technologies to integrate eight Web-based biological information sources for a sequence analysis service. With this service, users can ask a variety of queries about a given SNP (represented by a RS id, e.g., rs1799967). For example,

  1. Show me the sequence next to this SNP?
  2. How does this SNP probably affect protein structure?
  3. How many kind of transcription does the SNP lie in?

Answering these queries and the likes requires to access multiple Web-based bioinformatics resources. For example, we need to retrieve the sequence of the given SNP (i.e., a string of AGTC) from one database, and then submit this string to a set of Web sites that provide SNP function prediction services, and then compute the answer according to the results from different services. In this service, each information source is wrapped as a Web Service by a Web wrapper agent. Then we build an ontology of agents that represent the query answering power of each agent by specifying their input and output in RDF. With this ontology, we can compute a query answering workflow by a simple planning algorithm.




Koung-Lung Lin
Department of Computer Science and Information Engineering, National Taiwan University
Email: lkl@iis.sinica.edu.tw