New paths from dead ends

Modelling protein side-chain conformations using constraint logic programming

Side-chain placement is an important sub-task in protein modelling. Selecting conformations for side-chains is a difficult problem because of the large search space to be explored. This problem can be addressed using constraint logic programming (CLP), which is an artificial intelligence technique developed to solve large combinatorial search problems. The side-chain placement problem can be expressed as a CLP program in which rotamer conformations are used as values for finite domain variables, and bad steric contacts involving rotamers are represented as constraints. This paper introduces the concept of null rotamers, and shows how these can be used in implementing a novel iterative approach. We present results that compare the accuracy of models constructed using different rotamer libraries and different domain variable enumeration heuristics. The results obtained using this CLP-based approach compare favourably with those obtained by other methods.

Energy minimization method using automata network for sequence and side-chain conformation prediction from given backbone geometry

Globular proteins have high packing densities as a result of residue side chains in the core achieving a tight, complementary packing. The internal packing is considered the main determinant of native protein structure. From that point of view, we present here a method of energy minimization using an automata network to predict a set of amino acid sequences and their side-chain conformations from a desired backbone geometry for de novo design of proteins. Using discrete side-chain conformations, that is, rotamers, the sequence generation problem from a given backbone geometry becomes one of combinatorial problems. We focused on the residues composing the interior core region and predicted a set of amino acid sequences and their side-chain conformations only from a given backbone geometry. The kinds of residues were restricted to six hydrophobic amino acids (Ala, Ile, Met, Leu, Phe, and Val) because the core regions are almost always composed of hydrophobic residues. The obtained sequences were well packed as was the native sequence. The method can be used for automated sequence generation in the de novo design of proteins.

Combining evolutionary information and neural networks to predict protein secondary structure

Using evolutionary information contained in multiple sequence alignments as input to neural networks, secondary structure can be predicted at significantly increased accuracy. Here, we extend our previous three-level system of neural networks by using additional input information derived from multiple alignments. Using a position-specific conservation weight as part of the input increases performance. Using the number of insertions and deletions reduces the tendency for overprediction and increases overall accuracy. Addition of the global amino acid content yields a further improvement, mainly in predicting structural class. The final network system has sustained overall accuracy of 71.6% in a multiple cross-validation test on 126 unique protein chains. A test on a new set of 124 recently solved protein structures that have no significant sequence similarity to the learning set confirms the high level of accuracy. The average cross-validated accuracy for all 250 sequence-unique chains is above 72%. Using various data sets, the method is compared to alternative prediction methods, some of which also use multiple alignments: the performance advantage of the network system is at least 6 percentage points in three-state accuracy. In addition, the network estimates secondary structure content from multiple sequence alignments about as well as circular dichroism spectroscopy on a single protein and classifies 75% of the 250 proteins correctly into one of four protein structural classes. Of particular practical importance is the definition of a position-specific reliability index. For 40% of all residues the method has a sustained three-state accuracy of 88%, as high as the overall average for homology modelling. A further strength of the method is greatly increased accuracy in predicting the placement of secondary structure segments.

Redefining the goals of protein secondary structure prediction

Secondary structure prediction recently has surpassed the 70% level of average accuracy, evaluated on the single residue states helix, strand and loop (Q3). But the ultimate goal is reliable prediction of tertiary (three-dimensional, 3D) structure, not 100% single residue accuracy for secondary structure. A comparison of pairs of structurally homologous proteins with divergent sequences reveals that considerable variation in the position and length of secondary structure segments can be accommodated within the same 3D fold. It is therefore sufficient to predict the approximate location of helix, strand, turn and loop segments, provided they are compatible with the formation of 3D structure. Accordingly, we define here a measure of segment overlap (Sov) that is somewhat insensitive to small variations in secondary structure assignments. The new segment overlap measure ranges from an ignorance level of 37% (random protein pairs) via a current level of 72% for a prediction method based on sequence profile input to neural networks (PHD) to an average 90% level for homologous protein pairs. We conclude that the highest scores one can reasonably expect for secondary structure prediction are a single residue accuracy of Q3 > 85% and a fractional segment overlap of Sov > 90%.