The recently completed Protein Structure Initiative (PSI) resulted in over 13,400 Structural Genomics (SG) new protein structures in the Protein Data Bank (PDB). Many SG proteins have putative functional assignments but they are often incorrect. Our approach to functionally annotate SG proteins is based on local structure matching at the computationally-predicted active site. First, Partial Order Optimum Likelihood (POOL) is used to predict the catalytic residues of each protein within a given superfamily. Next, Structurally Aligned Local Sites of Activity (SALSA) is used to align and score the predicted catalytic residues of the well-characterized members of the superfamily. From this analysis, we generate chemical signatures for each functional family and compare them to the sets of catalytic residues predicted for each SG protein. In this project our focus is on the Crotonase Superfamily (CS), which consists of numerous, diverse functional families that are well characterized structurally and functionally and includes hydrolase, isomerase, hydratase, and dehalogenase activities. The CS also contains over 70 SG proteins, so it is ideal to test predictions of protein function. We demonstrate based on the above computational methods that the majority of the putative annotations in this superfamily are likely incorrect. Currently, biochemical assays are being used to test these predictions. For example, the SG protein <i>Thermus thermophilus </i>Q5SLS5_THET8 (PDB ID 1WZ8) was originally classified as a "probable enoyl-CoA hydratase". However, after computational analysis, we concluded that this protein more likely possesses hydrolase activity as the local active site of the SG protein does not align with the hydratase functional family, but aligns better with the hydrolase functional family. Biochemical results show that this SG protein possesses hydrolase activity, confirming our functional prediction. We have also determined that this SG protein interacts with the same substrate as the protein of known function (PDB ID 2J5S) and has significantly better catalytic efficiency when compared to this protein of known function. The main outcome of this project will be to classify successfully the biochemical functions of SG proteins within the CS based on their local structure at the predicted active sites. In addition, we will provide a validated approach to functional annotation for wider application by the community. (Support: NSF-CHE-1305655, PhRMA Foundation)

Utilizing computational chemistry to characterize the functions of structural genomics proteins

251st National Meeting (2016)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.