A significant fraction of proteins in all livings species, commonly known as intrinsically disordered proteins or IDPs, was considered to lack any well defined three-dimensional structure. IDPs were recognized as an important class of proteins for their functional importance. Recent studies suggested that naturally occurring sequences tend to be more disordered as compared to randomized sequences, suggesting that this feature is under evolutionary selection. However, the extent and intensity of selection in IDPs remain elusive. Moreover, there is no clear understanding about the regions those might be under evolutionary pressure for strong or weak disorder and their functional implications.
Here, in this study, based on the analysis of various model organisms (from both prokaryotic and eukaryotic domains) and viruses, we explored the signatures of selection for disordered residues in the respective proteins. Using different randomization models here we identified several regions with moderate to a strong preference for disordered residues when compared with random sequences. Specifically, we noticed evidence of selection for disordered residues both at the beginning and end of proteins in all the studied organisms. Further, here we showed that this preference is not due to the biases in nucleotide composition in those regions. Overall, in this study, we demonstrated for the first time a regional preference for disordered residues in both prokaryotic and eukaryotic proteins independent of the selection for nucleotide composition. The result may suggest that increased disorder in these regions contribute to the functionality of the relevant protein and to the organism fitness; however, the exact biophysical mechanisms related to this potentially improved functionality need to be further explored.