The continuous trend in computer science to lift programming to higher abstraction levels increases scalability and opens programming to a wider public. In particular, service-oriented programming and the support of semantics-based frameworks make application development accessible to users with almost no programming expertise. This monograph establishes requirement-centric scientific workflow design as an instance of consequent constraint-driven development. Requirements formulated in terms of user-level constraints are automatically transformed into running applications using temporal logic-based synthesis technology. The impact of this approach is illustrated by applying it to four very different bioinformatics scenarios: phylogenetic analysis, the dedicated GeneFisher-P scenario, the FiatFlux-P scenario, and microarray data analyses.