An Object-Oriented Programming System for the Integration of Internet-Based Bioinformatics Resources

The Internet consists of a vast inhomogeneous reservoir of data. Developing software that can integrate a wide variety of different data sources is a major challenge that must be addressed for the realisation of the full potential of the Internet as a scientific research tool. This article presents a semi-automated object-oriented programming system for integrating web-based resources. We demonstrate that the current Internet standards (HTML, CGI common gateway interface], Java, etc.) can be exploited to develop a data retrieval system that scans existing web interfaces and then uses a set of rules to generate new Java code that can automatically retrieve data from the Web. The validity of the software has been demonstrated by testing it on several biological databases. We also examine the current limitations of the Internet and discuss the need for the development of universal standards for web-based data.