There is a great interest to increase protein stability to enhance their utilization in biocatalysis, therapy, diagnostics, and production of nanomaterials. Experimental methods, like directed evolution, are costly and time-consuming, and therefore, computational methods present an attractive alternative. We have recently developed a method called FireProt for the design of stable proteins, which is available as a web server. In this thesis, the student will focus on i) preparation of a testing dataset of protein stability data, ii) testing different settings, iii) evaluation of different force-fields, and iv) combination of molecular dynamics simulation with stability evaluating tools as a potential alternative to the current approach. The optimal settings and method will be used as a new default on the FireProt web server.