The Identification Of A Simple Connector Protein Pdzk1 Ligand Of The Ligand Functional Interrelated To Predict And Verify The Two Through The Integration Of Machine Learning Algorithm To Predict The System Efficient Identification Of Hpv 16 E6 Interaction

Protein-protein interactions are intrinsic to virtually every cellular process. It is believed that different numbers or different species of ligands bind to different sites to facilitate the formation of different complexes, leading to different functions in different conditions. Adaptor protein play a critical role in the recruitment of protein complexes, it is important to life process.PDZK1 is a simple adaptor protein with four protein interaction PDZ domains, but without any other known functional domains. Here, we used yeast two-hybrid screening of a random peptide library and high-throughput validation screening of a specialized PDZ ligand candidate library to systematically and comprehensively identify PDZK1 ligands.The potential functional associations of the ligands were predicted by functional annotations from a MILANO literature search and subcellular localizations. We assumed that the ligands of an adaptor protein are more likely to be associated functionally if they meet one of the following criteria. Firstly, if two ligands of same protein have similar patterns of functions, they are likely to be functionally associated. Since the number of publications with co-occurrences of biological process terms roughly reflects the likelihood of their functional association in biological processes , two ligands were considered more likely to be associated functionally when both had similar functional patterns in their corresponding functional search for biological process terms. Secondly, if the ligands all have a closely related function, they are more likely to be functionally associated. The closely related functions could be defined by the ratio of co-occurrences of two biological processes in relevant publications.For some functionally associated ligand pairs, interaction with one ligand was found to be influenced by another ligand in a yeast three-hybrid system and we demonstrate D-AKAP2 disrupt the interaction of PDZK1 with BCR in mammalian cells. This strategy can be extended to the study of other adaptor proteins that contain peptide-binding domains.In the second chapter of this article, an integrated prediction system coupled with yeast two-hybrid confirmations was used to identify HPV 16 E6 interacting PDZ proteins. Identification of protein interactions by experimental screening is not only labor-intensive but almost futile in screening low abundance species due to the suppression by high abundance species. A better way is to use high-throughput computational prediction to screen protein sequence databases to produce manageable potential ligands and to validate experimentally. Here, we used an integrated machine learning system developed in the lab coupled with yeast two-hybrid conformations to identify Human papillomavirus (HPV) type 16 E6 interacting PDZ proteins. It has been reported that carcinogenic HPV 16 E6 has PDZ interacting C-terminal, which non-carcinogenic ones do not have. Using prediction plus validation strategy, VELI 3 was found to interact with HPV 16 E6 in Y2H system, suggest that VELI3 is a potential cellular target of HPV 16 E6. This strategy can be easily extended to a variety of viral proteins with PDZ binding motif for identification of their targeting cellular PDZ domain proteins.