just wondering what type of thing would go into this topic
Scot E. DOwd Ph.D. wrote:
I think that a "workflow" or a "pipeline" is a more frequently used term for a computational assay. A simple workflow would be 1) retrieve DNA sequence from database, 2) run Blast against another database, 3) return results. Obviously things can get much more complex.
A major problem in bioinformatics is the frequent lack of rigour in computational assay development - when biologists develop workflows that involve manually moving information from one analytical or data lookup service to the next, it becomes difficult to keep track of all the steps and difficult to reproduce the experiments (especially as the database and analytical services often change). The MyGrid project and the BioMoby project both seek to make it easier to build, reproduce and monitor the execution of complex, distributed workflows.
A good example of a complicated bioinformatics assay that produced real world, useful results is described in this paper about Williams-Beuren syndrome
More information on workflow development can be found below.