In modern scientific researches, a lot of computer programs have been developed for scientists to do different computation or analysis on their data. And database technologies are commonly used in e-Science community to store massive mount of scientific data. But the challenge still exists.
In terms of computation, the existing tools usually don't have a user-friendly interface and require installation in the users' machines. And using these tools to do computation may be too slow in a personal computer. In terms of data, some of the tools are quite strict on the data format which is not always available to the users. A new solution is needed.
I will discuss an e-Science solution of a real world scientific problem in
bioinformatics. It is a prototype system which organizes existing command line programs into a pipeline architecture and provides a web interface for users to do analysis remotely. On top of an interface what user prefers, the command line tools is much more usable to the users and the data format transformation is done along the pipeline automatically behind the scene. And provenance information is also recorded for each experiment users perform within the system.