Monte Carlo (MC) path generation and small-world network approach to identify functional residues in proteins

Abstract

Allosteric mechanism of proteins is crucial for their proper functioning. An important aspect underlying this mechanism is the communication pathways connecting the regulatory and active sites of proteins. Monte Carlo (MC) path generation is proposed as a novel methodology to identify the ensemble of these pathways and the residues taking part in the communication. This methodology is applied to determine the allosteric communication pathways in tetramerization domain of the Shaker potassium channel, PSD-95, bovine rhodopsin, and Dictyostelium myosin II. Another aspect derived from MC path generation method is generating a network of residues and finding the functional residues by applying the small-world network approach. The network parameters such as betweenness, closeness, and clustering coefficient are calculated to be used as tools for predicting the functionally important residues. In addition to the mentioned systems, this approach is also applied to the HIV-1 protease system. The suggested pathways and the residues constituting these pathways are mostly in agreement with the previous studies in literature. Also, the majority of the residues proposed as functionally important are consistent with previous studies. Furthermore, the small-world network approach is applied to the HIV-1 protease system; the network parameters are calculated for both mutant and co-evolved structures and compared with the wild-type structure. The highest correlation is observed between the wild-type and the co-evolved structures, which justifies the existence of this mutation for the conservation of network properties. Overall, the proposed computational methodology has the potential to identify the residues mediating the allosteric communication and to detect the functionally important sites in proteins.