Mario Compiani

Predictions of protein segments with the same aminoacid sequence and different secondary structure: a benchmark for predictive methods.

The most stringent test for predictive methods of protein secondary structure is whether identical short sequences that are known to be present with different conformations in different proteins known at atomic resolution can be correctly discriminated. In this study, we show that the prediction efficiency of this type of segments in unrelated proteins reaches an average accuracy per residue ranging from about 72 to 75% (depending on the alignment method used to generate the input sequence profile) only when methods of the third generation are used. A comparison of different methods based on segment statistics (2nd generation methods) and/or including also evolutionary information (3rd generation methods) indicate that the discrimination of the different conformations of identical segments is dependent on the method used for the prediction. Accuracy is similar when methods similarly performing on the secondary structure prediction are tested. When evolutionary information is taken into account as compared to single sequence input, the number of correctly discriminated pairs is increased twofold. The results also highlight the predictive capability of neural networks for identical segments whose conformation differs in different proteins. Proteins 2000;41:535–544.

Predicting secondary structures of membrane proteins with neural networks.

Back-propagation, feed-forward neural networks are used to predict the secondary structures of membrane proteins whose structures are known to atomic resolution. These networks are trained on globular proteins and can predict globular protein structures having no homology to those of the training set with correlation coefficients (C) of 0.45, 0.32 and 0.43 for αa-helix, β-strand and random coil structures, respectively. When tested on membrane proteins, neural networks trained on globular proteins do, on average, correctly predict (Qi) 62%, 38% and 69% of the residues in the α-helix, β-strand and random coil structures. These scores rank higher than those obtained with the currently used statistical methods and are comparable to those obtained with the joint approaches tested so far on membrane proteins. The lower success score for β-strand as compared to the other structures suggests that the sample of β-strand patterns contained in the training set is less representative than those of a-helix and random coil. Our analysis, which includes the effects of the network parameters and of the structural composition of the training set on the prediction, shows that regular patterns of secondary structures can be successfully extrapolated from globular to membrane proteins.

A PREDICTOR OF TRANSMEMBRANE ALPHA -HELIX DOMAINS OF PROTEINS BASED ON NEURAL NETWORKS

Back-propagation, feed-forward neural networks are used to predict a-helical transmembrane segments of proteins. The networks are trained on the few membrane proteins whose transmembrane α-helix domains are known to atomic or nearly atomic resolution. When testing is performed with a jackknife procedure on the proteins of the training set, the fraction of total correct assignments is as high as 0.87, with an average length for the transmembrane segments of 20 residues. The method correctly fails to predict any transmembrane domain for porin, whose transmembrane segments are β-sheets. When tested on globular proteins, lower and upper limits of 1.6 and 3.5% for a total of 26826 residues are determined for the mispredicted cases, indicating that the predictor is highly specific for α-helical domains of membrane proteins. The predictor is also tested on 37 membrane proteins whose transmembrane topology is partially known. The overall accuracy is 0.90, two percentage points higher than that obtained with statistical methods. The reliability of the prediction is 100% for 60% of the total 18242 predicted residues of membrane proteins. Our results show that the local directional information automatically extracted by the neural networks during the training phase plays a key role in determining the accuracy of the prediction.

LGANN: a parallel system combining a local genetic algorithm and neural networks for the prediction of secondary structure of proteins

In this work we describe a parallel system consisting of feed-forward neural networks supervised by a local genetic algorithm. The system is implemented in a transputer architecture and is used to predict the secondary structures of globular proteins. This method allows a wide search in the parameter space of the neural networks and the determination of their optimal topology for the predictive task. Different neural network topologies are selected by the genetic algorithm on the basis of minimal values of mean square errors on the testing set. When the alpha-helix, beta-strand and random coil motifs of secondary structures are discriminated, the maximal efficiency obtained is 0.62, with correlation coefficients of 0.35, 0.31 and 0.37 respectively. This level of accuracy is similar to that previously attained by means of neural networks without hidden layers and using single protein sequences as input. The results validate the neural network topologies used for the prediction of protein secondary structures and highlight the relevance of the input information in determining the limit of their performance.

Computational and Theoretical Methods for Protein Folding

A computational approach is essential whenever the complexity of the process under study is such that direct theoretical or experimental approaches are not viable. This is the case for protein folding, for which a significant amount of data are being collected. This paper reports on the essential role of in silico methods and the unprecedented interplay of computational and theoretical approaches, which is a defining point of the interdisciplinary investigations of the protein folding process. Besides giving an overview of the available computational methods and tools, we argue that computation plays not merely an ancillary role but has a more constructive function in that computational work may precede theory and experiments. More precisely, computation can provide the primary conceptual clues to inspire subsequent theoretical and experimental work even in a case where no preexisting evidence or theoretical frameworks are available. This is cogently manifested in the application of machine learning methods to come to grips with the folding dynamics. These close relationships suggested complementing the review of computational methods within the appropriate theoretical context to provide a self-contained outlook of the basic concepts that have converged into a unified description of folding and have grown in a synergic relationship with their computational counterpart. Finally, the advantages and limitations of current computational methodologies are discussed to show how the smart analysis of large amounts of data and the development of more effective algorithms can improve our understanding of protein folding.

Escape rates in bistable systems with position‐dependent friction coefficients

In this paper we consider the generalization of the Kramers’ model of chemical reactions to the case that the friction coefficient γ(x) depends on the reaction coordinate x. Extending previous efforts the escape rate is exactly evaluated in the high‐friction limit imposing on γ(x) much milder conditions than used so far in the literature. The ensuing rate retains the Kramers’ form and reproduces the renormalization effect of the damping coefficient which has been reported in laboratory experiments on chemical reactions. The origin of variable friction coefficients is then discussed within the framework of a multidimensional Markovian model and ascribed to the nonlinear coupling of the reaction coordinate with faster auxiliary variables. Finally, the implications of our results for the rate of ligands migration in proteins are briefly considered.

Reduction of Active Elastase Concentration by Means of Immobilized Inhibitors: A Novel Therapeutic Approach

The inhibitory power of three different active Nylon membranes, separately loaded with three different protease inhibitors, was studied with the aim of reducing the increased elastase concentration occurring during hemodialysis or extracorporeal blood circulation in patients undergoing cardiopulmonary bypass. Chemical grafting was carried out to make the inert Nylon membrane suitable for the immobilization of the inhibitors. The behavior of immobilized α1‐antitrypsin, bovine pancreatic trypsin inhibitor (BPTI), or elastatinal was separately studied. α1‐Antitrypsin and BPTI were covalently immobilized by means of a diazotization process, whereas elastatinal was covalently attached via a condensation process mediated by glutaraldehyde. The inhibitory power of each membrane type was studied as a function of the amount of immobilized inhibitor and temperature. All active membranes have shown good inhibitory power. The most efficient membrane was that loaded with α1‐antitrypsin, the less efficient that with BPTI.

Neural networks predict protein folding and structure: artificial intelligence faces biomolecular complexity.

Abstract In the genomic era DNA sequencing is increasing our knowledge of the molecular structure of genetic codes from bacteria to man at a hyperbolic rate. Billions of nucleotides and millions of aminoacids are already filling the electronic files of the data bases presently available, which contain a tremendous amount of information on the most biologically relevant macromolecules, such as DNA. RNA and proteins. The most urgent problem originates from the need to single out the relevant information amidst a wealth of general features. Intelligent tools are therefore needed to optimise the search. Data mining for sequence analysis in biotechnology has been substantially aided by the development of new powerful methods borrowed from the machine learning approach. In this paper we discuss the application of artificial feedforward neural networks to deal with some fundamental problems tied with the folding process and the structure-function relationship in proteins.

Protein structure prediction and biomolecular recognition: From protein sequence to peptidomimetic design with the human β 3 integrin

Computational tools can bridge the gap between sequence and protein 3D structure based on the notion that information is to be retrieved from the databases and that knowledge-based methods can help in approaching a solution of the protein-folding problem. To this aim our group has implemented neural network-based predictors capable of performing with some success in different tasks, including predictions of the secondary structure of globular and membrane proteins, the topology of membrane proteins and porins and stable f -helical segments suited for protein design. Moreover we have developed methods for predicting contact maps in proteins and the probability of finding a cysteine in a disulfide bridge, tools which can contribute to the goal of predicting the 3D structure starting from the sequence (the so called ab initio prediction). All our predictors take advantage of evolution information derived from the structural alignments of homologous (evolutionary related) proteins and taken from the sequence and structure databases. When it is necessary to build models for proteins of unknown spatial structure, which have very little homology with other proteins of known structure, non-standard techniques need to be developed and the tools for protein structure predictions may help in protein modeling. The results of a recent simulation performed in our lab highlights the role of high performing computing technology and of tools of computational biology in protein modeling and peptidomimetic design.

Diffusion-Collision of Foldons Elucidates the Kinetic Effects of Point Mutations and Suggests Control Strategies of the Folding Process of Helical Proteins

In this article we use mutation studies as a benchmark for a minimal model of the folding process of helical proteins. The model ascribes a pivotal role to the collisional dynamics of a few crucial residues (foldons) and predicts the folding rates by exploiting information drawn from the protein sequence. We show that our model rationalizes the effects of point mutations on the kinetics of folding. The folding times of two proteins and their mutants are predicted. Stability and location of foldons have a critical role as the determinants of protein folding. This allows us to elucidate two main mechanisms for the kinetic effects of mutations. First, it turns out that the mutations eliciting the most notable effects alter protein stability through stabilization or destabilization of the foldons. Secondly, the folding rate is affected via a modification of the foldon topology by those mutations that lead to the birth or death of foldons. The few mispredicted folding rates of some mutants hint at the limits of the current version of the folding model proposed in the present article. The performance of our folding model declines in case the mutated residues are subject to strong long‐range forces. That foldons are the critical targets of mutation studies has notable implications for design strategies and is of particular interest to address the issue of the kinetic regulation of single proteins in the general context of the overall dynamics of the interactome. Proteins 2006.