Protein three-dimensional structure prediction

Abstract

This article deals with protein three-dimensional structure prediction, and focuses on how to make structural sense out of a single input sequence with unknown structure, the ‘query’ or ‘target’ sequence. Both template based and template free modelling techniques are treated, and how resulting structural models may be selected and refined. We also discuss the way in which structural predictions are validated within the global community, and elaborate on the extent to which predicted structures may be trusted and used in practice. We give a concrete flowchart for how to decide which modelling strategy is best suited in particular circumstances, and which steps need to be taken in each strategy. Notably, the ability to locate a suitable structural template by homology or fold recognition is crucial; without this models will be of low quality at best. With a template available, the quality of the query-template alignment crucially determines the model quality. We also discuss how other, courser, experimental data may be incorporated in the modelling process to alleviate the problem of missing template structures. Finally we discuss whether the concept of a single fold pertaining to a protein structure is sustainable given recent insights. In short, we conclude that the general protein three-dimensional structure prediction problem remains unsolved, especially if we desire quantitative predictions. However, if a homologous structural template is available in the PDB model or reasonable to high accuracy may be generated.