This case study describes research experience in orchestrating scientific, grid-based workflows using BPEL. The study orchestrates Grid services for the automation of a polymorph prediction application. Using this example, the paper explains the extent with which the BPEL language supports the definition of scientific workflows. It describes the reliability, performance and scalability that can be achieved by executing a complex scientific workflow with ActiveBPEL, an industrial strength but freely available BPEL engine.