Specifying and managing bioinformatics studies is becoming more commonplace via the use of scientific workflow management systems. Bioinformaticians like their programming paradigm because it allows them to quickly construct elaborate data processing pipelines. A graph structure forms the basis of this kind of model, with nodes standing in for individual bioinformatics activities and connections for the flow of information between them. There may be consequences for the reusability of scientific operations when the complexity of such network structures grows over time. In this paper, we advocate for the Taverna model as a means to efficiently design workflows. We contend that "anti-patterns," a word often used in program design, are a major cause of the problems associated with reuse since they imply the usage of idiomatic forms that result in too intricate design. This work's key contribution is a mechanism for automatically identifying such anti-patterns and replacing them with alternative patterns that reduce the structural complexity of the process. This approach to rewriting routines will improve operational efficiency while also improving the user experience (via simpler design and maintenance). (Easier to manage, and sometimes to exploit the latent parallelism amongst the tasks).