Home > Protein structure prediction
Protein structure prediction is one of the most significant tasks tackled in computational structural biology. It has the aim of determining the three-dimensional structure of proteins from their amino acid sequences. In more formal terms, this is the prediction of protein tertiary structure from primary structure. Given the usefulness of known protein structures in such valuable tasks as rational drug design, this is a highly active field of research.Every two years, the performance of current methods is assessed in the CASP experiment.
1 Overview
The practical role of protein structure prediction is now more important than ever. Massive amounts of protein sequence data may be derived from modern large-scale DNA sequencing efforts of, for example, the Human Genome Project. The output of experimentally determined protein structures, typically by time-consuming and relatively expensive X-ray crystallography or NMR spectroscopy, is lagging far behind the output of protein sequences.
A number of factors exist that make protein structure prediction a very difficult task, including:
- The number of possible structures that proteins may possess is extremely large, as highlighted by the Levinthal paradox.
- The physical basis of protein structural stability is not fully understood.
- The primary sequence may not fully specify the tertiary structure. For example, proteins known as chaperonins have the ability to induce proteins to fold in specific ways.
- Direct simulation of protein folding via methods such as molecular dynamicsMolecular dynamics (MD) simulation numerically solves Newton's equations of motion on an atomistic or similar model of a molecular system to obtain information about its time-dependent properties. Applications Beginning in theoretical physics, the method is not generally tractable for both practical and theoretical reasons.
Despite the above hinderances, much progress is being made by the many research groups that are interested in the task. Prediction of structures for small proteins is now a perfectly realistic goal. A wide range of approaches are routinely applied for such predictions. These approaches may be classified into two broad classes; de novo modelling and comparative modelling.
2 De novo protein modelling
De novo- or ab initioThe Latin term ab initio means from the beginning and is used in several contexts: when describing literature: told from the beginning as opposed to in medias res (meaning starting in the middle of the story . See also: List of Latin phrases as a legal te- protein modelling methods seek to build three-dimensional protein models "from scratch". There are many possible procedures that either attempt to mimic protein folding or apply some stochasticStochastic from the Greek "stochos" or "goal", means of, relating to, or characterized by conjecture; conjectural; random. A stochastic process is one whose behavior is non-deterministic in that the next state of the environment is not fully determined by method to search possible solutions (i.e. global optimizationGlobal optimization is a branch of applied mathematics and numerics that deals with the optimization of a function or a set of functions to some criteria. General The most common form is the minimization of one real-valued function in the parameter-space. of a suitable energy function). These procedures tend to require vast computational resources, and have thus only been carried out for tiny proteins. To attempt to predict protein structure de novo for larger proteins, we will need better algorithms and larger computational resources like those afforded by either powerful supercomputers (such as Blue GeneBlue Gene refers to an IBM project intended to develop the next class of petaflop-scale supercomputers. The first computer in the Blue Gene series, Blue Gene/L developed through a partnership with Lawrence Livermore National Laboratory, cost $100 million when it goes online) or distributed computing (see folding at homeFolding at home (or Folding@home is a distributed computing project designed to perform computationally intensive simulations of protein folding. It was launched on October 1, 2000, and is currently managed by the Pande Group, within Stanford University's). Although these computational barriers are vast the potential benifits of structural genomics (by predicted or experimental methods) make de novo structure prediction an active research field.
