Towards Fully Automated Structure-based Function Prediction in Structural Genomics: A Case Study

Abstract

As the global Structural Genomics projects have picked up pace, the number of structures annotated in the Protein Data Bank as hypothetical protein or unknown function has grown significantly. A major challenge now involves the development of computational methods to assign functions to these proteins accurately and automatically. As part of the Midwest Center for Structural Genomics (MCSG) we have developed a fully automated functional analysis server, ProFunc, which performs a battery of analyses on a submitted structure. The analyses combine a number of sequence-based and structure-based methods to identify functional clues. After the first stage of the Protein Structure Initiative (PSI), we review the success of the pipeline and the importance of structure-based function prediction. As a dataset, we have chosen all structures solved by the MCSG during the 5 years of the first PSI. Our analysis suggests that two of the structure-based methods are particularly successful and provide examples of local similarity that is difficult to identify using current sequence-based methods. No one method is successful in all cases, so, through the use of a number of complementary sequence and structural approaches, the ProFunc server increases the chances that at least one method will find a significant hit that can help elucidate function. Manual assessment of the results is a time-consuming process and subject to individual interpretation and human error. We present a method based on the Gene Ontology (GO) schema using GO-slims that can allow the automated assessment of hits with a success rate approaching that of expert manual assessment.

Maps of protein structure space reveal a fundamental relationship between protein structure and function

“…The level of detail in the GO-MF graph is uneven: Some areas are better studied and correspond to subgraphs of the GO-MF graph that have more levels and, ultimately, more functional annotations. In addition, proteins of the same function sometimes have annotations at different levels (14). One could argue that perhaps the proteins that lie in the core happen to have functions that are described in finer detail, and the apparent high diversity of this core is merely an artifact of the uneven level of detail in the GO-MF graph.…”

Systematic Structural Characterization of Metabolites in Arabidopsis via Candidate Substrate-Product Pair Networks

“…in Table 1). To obtain all nodes that represent compounds of the same biochemical class, the same strategy was followed as used in automated gene and protein function annotation (Watson et al, 2007;Loewenstein et al, 2009;Klie et al, 2012). Given the CSPP network, first a subnetwork on nodes representing known compounds of the biochemical class of interest was extracted (together with the incident edges).…”

Exploring the structure and function paradigm

“…Although the majority of structural data are associated with additional experimental data regarding a protein's function, this is not the case for a large proportion of PSI structures [3]. For these proteins it is necessary to attempt to predict the function from the structure.…”

“…By boosting the structural repertoire, it is hoped that we will be able to improve our understanding of fold space and how proteins evolve new functions. Consequently, the four major PSI Centres have targeted large sequence families that are most likely to adopt novel structures [1] (http://www.structuralgenomics.org), although these often have little or no functional annotation [3]. To maximise the biomedical benefit of PSI structures, recent reviews have proposed broadening selection criteria to explicitly focus on the relevance to human disease [4] or to provide structural characterisation of families with known functions [5].…”