Recombinant protein expression in bacteria often results in the formation of both inactive and insoluble protein that accumulates as intracellularprotein aggregates called inclusion bodies [1]. It has beenshown that 70-80% of recombinant proteins expressed in E.coli are as inclusion bodies [2]. This is probably due to the independence of the protein type in bacterial systems. In cases of expression of eukaryotic proteins, which usually contain cysteines that are prone to form disulfide bonds in the nativestate, the bacterial system maynot support the appropriate pairing of disulfide bonds in the newly-produced protein thus leads to the presence of insoluble protein pellets [3].

The three-dimensional structure of a protein is essential to the understanding of its biological function and the design of drug candidates. In recent years, tremendous progress has been achieved in structural biology and a large of number of new macromolecular structures have been determined using experimental methods such as X-ray crystallography, NMR and cryo-EM. Despite the impressive advances, it is still generally difficult and time-consuming to obtain structures for every protein of interest. This is why computational approaches come into play as a powerful alternative and supplement to experimental methods.