Using neural networks and evolutionary information in decoy discrimination for protein tertiary structure prediction

Background  

We present a novel method of protein fold decoy discrimination using machine learning, more specifically using neural networks. Here, decoy discrimination is represented as a machine learning problem, where neural networks are used to learn the native-like features of protein structures using a set of positive and negative training examples. A set of native protein structures provides the positive training examples, while negative training examples are simulated decoy structures obtained by reversing the sequences of native structures. Various features are extracted from the training dataset of positive and negative examples and used as inputs to the neural networks.     

Negative Example Selection for Protein Function Prediction: The NoGO Database

Negative examples – genes that are known not to carry out a given protein function – are rarely recorded in genome and proteome annotation databases, such as the Gene Ontology database. Negative examples are required, however, for several of the most powerful machine learning methods for integrative protein function prediction. Most protein function prediction efforts have relied on a variety of heuristics for the choice of negative examples. Determining the accuracy of methods for negative example prediction is itself a non-trivial task, given that the Open World Assumption as applied to gene annotations rules out many traditional validation metrics. We present a rigorous comparison of these heuristics, utilizing a temporal holdout, and a novel evaluation strategy for negative examples. We add to this comparison several algorithms adapted from Positive-Unlabeled learning scenarios in text-classification, which are the current state of the art methods for generating negative examples in low-density annotation contexts. Lastly, we present two novel algorithms of our own construction, one based on empirical conditional probability, and the other using topic modeling applied to genes and annotations. We demonstrate that our algorithms achieve significantly fewer incorrect negative example predictions than the current state of the art, using multiple benchmarks covering multiple organisms. Our methods may be applied to generate negative examples for any type of method that deals with protein function, and to this end we provide a database of negative examples in several well-studied organisms, for general use (The NoGO database, available at: bonneaulab.bio.nyu.edu/nogo.html).     

Effect of enzyme-enzyme interactions on steady-state enzyme kinetics. IV. “Strictly steady-state” examples

A number of two-state and three-state examples of enzyme interaction effects at steady state are considered. In contrast to most examples in parts I, II, and III, these are not of the quasi-equilibrium type. The Bragg—Williams approximation is used here for lattices of both two-state and three-state enzymes. In addition, several examples of small (oligomeric) systems are treated. Diffusion in lattice-fluid models is introduced in a simple way in the text and commented on further in the appendix (together with diffusion in solution).     

Using Protistan Examples to Dispel the Myths of Intelligent Design

ABSTRACT. In recent years the teaching of the religiously based philosophy of intelligent design (ID) has been proposed as an alternative to modern evolutionary theory. Advocates of ID are largely motivated by their opposition to naturalistic explanations of biological diversity, in accordance with their goal of challenging the philosophy of scientific materialism. Intelligent design has been embraced by a wide variety of creationists who promote highly questionable claims that purport to show the inadequacy of evolutionary theory, which they consider to be a threat to a theistic worldview. We find that examples from protistan biology are well suited for providing evidence of many key evolutionary concepts, and have often been misrepresented or roundly ignored by ID advocates. These include examples of adaptations and radiations that are said to be statistically impossible, as well as examples of speciation both in the laboratory and as documented in the fossil record. Because many biologists may not be familiar with the richness of the protist evolution dataset or with ID‐based criticisms of evolution, we provide examples of current ID arguments and specific protistan counter‐examples.     

Spontaneous oscillations in two 2-component cells coupled by diffusion

Mathematical examples are presented of oscillators with two variables which do not oscillate in isolation, but which do oscillate stably when coupled with a twin via difiusion. Two examples are presented, the LefeverPrigogine Brusselator and a system used to model glycolytic oscillations. The mathematical method is not the usual bifurcation theory, but rather a type of singular perturbation theory combined with bifurcation theory. For both examples, it is shown that all stationary solutions are unstable for appropriate parameter settings. In the case of the Brusselator, it is further shown that there exist limit cycles; i.e. stable oscillations, in this parameter range. A numerical example is presented.Partially supported by NSF     

Economic allocation: Examples and derived decision tree

Goal, Scope and Background  In the recently published (Dutch) Handbook on LCA, economic allocation is advised as baseline method for most allocation situations in a detailed LCA. Although the Handbook on LCA aimed to provide a ‘cookbook’ with operational guidelines for conducting each step of an LCA, this was not completely achieved for the allocation step. The guidelines for allocation largely remained at the level of principles. This restricted elaboration of economic allocation may hamper application in practice. Therefore, this paper elaborates some examples applying economic allocation. Method  Two concepts are of particular importance when applying economic allocation: functional flow and multi-functional process. The definitions of these concepts are presented and discussed. The basic principle of economic allocation is that having determined the various functional flows of a multi-functional process, all other flows need to be allocated to these functional flows according to their shares in the total proceeds. Proceeds are based on prices and these are not always easy to determine for a process. A summary of possible solutions for different problems when determining prices is given. Results and Discussion  The examples presented focus on co-production and various recycling situations. All examples are hypothetical in order to avoid discussions on the data. The examples show that the prices of the functional flows determine the allocation results. It is of importance to have correct information on the relative prices of the functional flows at stake, especially whether they are negative or positive. Learning from these examples, we establish a decision tree for economic allocation. The decision tree is meant for identifying and handling multi-functionality situations starting from a defined (product) system. This decision tree is with minor adaptations also applicable to other allocation methods and has a more general value than for the economic allocation method only. Conclusions and perspective  The examples have helped us to establish a decision tree for handling the multi-functionality problem by economic allocation. The examples can be broadened to other materials and allocation situations. We would encourage others to provide other examples and experiences as we expect that these will help to further improve and refine the guidelines and decision tree for economic allocation in future.     

Artificial neural networks and their use in quantitative pathology

A brief general introduction to artificial neural networks is presented, examining in detail the structure and operation of a prototype net developed for the solution of a simple pattern recognition problem in quantitative pathology. The process by which a neural network learns through example and gradually embodies its knowledge as a distributed representation is discussed, using this example. The application of neurocomputer technology to problems in quantitative pathology is explored, using real-world and illustrative examples. Included are examples of the use of artificial neural networks for pattern recognition, database analysis and machine vision. In the context of these examples, characteristics of neural nets, such as their ability to tolerate ambiguous, noisy and spurious data and spontaneously generalize from known examples to handle unfamiliar cases, are examined. Finally, the strengths and deficiencies of a connectionist approach are compared to those of traditional symbolic expert system methodology. It is concluded that artificial neural networks, used in conjunction with other nonalgorithmic artificial intelligence techniques and traditional algorithmic processing, may provide useful software engineering tools for the development of systems in quantitative pathology.     

A study of the 8466 structures reported in Inorganica Chimica Acta: 52 space group changes and their chemical consequences

260 structures published in Inorganica Chimica Acta present some problem about their crystal and molecular symmetry, moreover 41 compounds have been already corrected in this journal or in other journals. The unavailability of the structure factors tables has prevented a complete correction, in fact, we have reached a definitive result only for 52 crystalline compounds. These are divided into five categories: (A) incorrect Laue Group (19 examples), (B) omission of a centre of symmetry (20 examples), (C) incorrect Laue group and omission of a centre of symmetry (one example) (D) omission of a centre of symmetry coupled with a failure to recognize systematic absences (10 examples) and finally (E) non-space group translations (two examples). Moreover, other 44 examples of non-space group translations might be present, but we have not reached a definitive result for the unavailability of the structure factor tables.     

Introductions and extinction of fish in the African great lakes

The great lakes of Africa contain one of the most remarkable known examples of rapid evolution and speciation of a vertebrate group. The three major lakes - Lake Victoria, Lake Malawi and Lake Tanganyika - each contain a unique radiation of fish belonging to the family Cichlidae. This has produced species 'flocks' that are unique to each lake in species numbers and diversity. At present most of the cichlid species in Lake Victoria are facing extinction as a result of the introduction of an exotic fish species. There are many other examples of introductions or invasions of non-native fish that have had severe effects on the native fish populations and regional economics. These examples should provide a clear warning against the proposed introductions of exotic fish into Lake Malawi.     

Quantum-like interference effect in gene expression: glucose-lactose destructive interference

In this note we illustrate on a few examples of cells and proteins behavior that microscopic biological systems can exhibit a complex probabilistic behavior which cannot be described by classical probabilistic dynamics. These examples support authors conjecture that behavior of microscopic biological systems can be described by quantum-like models, i.e., models inspired by quantum-mechanics. At the same time we do not couple quantum-like behavior with quantum physical processes in bio-systems. We present arguments that such a behavior can be induced by information complexity of even smallest bio-systems, their adaptivity to context changes. Although our examples of the quantum-like behavior are rather simple (lactose-glucose interference in E. coli growth, interference effect for differentiation of tooth stem cell induced by the presence of mesenchymal cell, interference in behavior of PrP(C) and PrP(Sc) prions), these examples may stimulate the interest in systems biology to quantum-like models of adaptive dynamics and lead to more complex examples of nonclassical probabilistic behavior in molecular biology.     

Numerical experiments with a new differentiation filter

A dynamic programming filter which provides estimates of the first and second derivative of empirical displacement data is investigated numerically. This filter uses a weighted least squares criteria in estimating the derivatives. The filter equations are presented together with several numerical examples. These examples are taken from references that proposed other techniques.     

The human THE-LTR(O) and MstII interspersed repeats are subfamilies of a single widely distributed highly variable repeat family.

Fifteen examples of the transposon-like human element (THE) LTR and thirteen examples of the MstII interspersed repeat are aligned to generate new consensus sequences for these human repetitive elements. The consensus sequences of these elements are very similar, indicating that they compose subfamilies of a single human interspersed repetitive sequence family. Members of this highly polymorphic repeat family have been mapped to at least 11 chromosomes. Seven examples of the THE internal sequence are also aligned to generate a new consensus sequence for this element. Estimates of the abundance of this repetitive sequence family, derived from both hybridization analysis and frequency of occurrence in GenBank, indicate that THE-LTR/MstII sequences are present every 100-3000 kb in human DNA. The widespread occurrence of members of this family makes them useful landmarks, like Alu, L1, and (GT)n repeats, for physical and genetic mapping of human DNA.     

Stochastic equilibria for multivariate reaction systems

The equilibrium distribution for a generalQth-order multivariate reaction system is studied. The state transition intensity matrix is developed and examples are given for small numbers of reaction components. A closed-form expression for the equilibrium distribution for systems which are symmetric with respect to the order of component reactions is presented. Numerical examples for three component systems are discussed.     

Diskontinuierlich aktive Gene und Evolution Eine Diskussion am Beispiel der Archaeognatha (Insecta)

Discontinuous activity of genes and evolution - a discussion based on examples of the Archaeognatha (Insecta) On the basis of examples, mainly from the group of Archaeognatha (Insecta) it seems probable that, in multicellular organisms, there is not only a discontinuous activity of genes and groups of genes during the individual development, but also during the phylogenesis. The possibility that silent genes can be reactivated is probably an essential reason for the existence of parallelisms and homologous tendencies. Both conceptions have a complementary importance for an establishment of taxa ana the reconstruction of phylogenesis and are defined according to evolutionary genetics. The examples are discussed and the necessity of similar comparative studies in other supraspecific taxa is emphasized.     

The curious case of protein splicing: mechanistic insights suggested by protein semisynthesis.

The gradual accumulation of examples of protein splicing, in which a nested intervening sequence is spliced out of the interior of a polyprotein precursor, suggests that this curious phenomenon might prove to have universal phylogenetic distribution and biological significance. The known examples are reviewed, with the aim of establishing underlying patterns, and a generalized mechanism of autocatalytic protein splicing is proposed. The testable consequences of such a proposal and the possible evolutionary origins of the phenomenon are discussed.     

Predicting gene function using few positive examples and unlabeled ones

Background

A large amount of functional genomic data have provided enough knowledge in predicting gene function computationally, which uses known functional annotations and relationship between unknown genes and known ones to map unknown genes to GO functional terms. The prediction procedure is usually formulated as binary classification problem. Training binary classifier needs both positive examples and negative ones that have almost the same size. However, from various annotation database, we can only obtain few positive genes annotation for most offunctional terms, that is, there are only few positive examples for training classifier, which makes predicting directly gene function infeasible.

Results

We propose a novel approach SPE_RNE to train classifier for each functional term. Firstly, positive examples set is enlarged by creating synthetic positive examples. Secondly, representative negative examples are selected by training SVM(support vector machine) iteratively to move classification hyperplane to a appropriate place. Lastly, an optimal SVM classifier are trained by using grid search technique. On combined kernel ofYeast protein sequence, microarray expression, protein-protein interaction and GO functional annotation data, we compare SPE_RNE with other three typical methods in three classical performance measures recall R, precise P and their combination F: twoclass considers all unlabeled genes as negative examples, twoclassbal selects randomly same number negative examples from unlabeled gene, PSoL selects a negative examples set that are far from positive examples and far from each other.

Conclusions

In test data and unknown genes data, we compute average and variant of measure F. The experiments showthat our approach has better generalized performance and practical prediction capacity. In addition, our method can also be used for other organisms such as human.

     

Role of basic science in the development of new medicines: examples from the eicosanoid field

The role of basic science in the development of health care has received more and more attention. In my own area of research involving the so-called eicosanoids, there are many examples of how studies of structure and function of small molecules, as well as proteins and genes, have led to new therapeutic agents for treatment of a variety of diseases. In most of the cases, the discoveries have resulted in the recognition of novel therapeutic targets amenable to modulation by small molecules. However, there are also examples in which the molecular mechanisms of actions of drugs, discovered by phenotypic screening, have been elucidated. The majority of the examples in this article consist of approved drugs; however, in some cases, ongoing developments of potential therapeutics are cited.     

Multistep reactions with immobilized microorganisms

This review presents some examples of new semicontinuous and continuous processes for product formation and for degradation of xenobiotics with immobilized microorganisms. A semicontinuous process for glycerol formation with Saccharomyces cerevisiae cells, adsorbed on sintered glass, and a continuous production of citric acid with Aspergillus niger, entrapped in calcium alginate, are given as examples of the production of primary metabolites. The production of ergot alkaloids with entrapped Claviceps purpurea demonstrates the possibilities of production of secondary metabolites. The continuous degradation of phenolic substances and that of 4-chlorophenol by entrapped and adsorbed microorganisms are given as examples of the possibility of a continuous degradation of xenobiotics by immobilized microorganisms. The continuous degradation of these substances was successful not only in artificial solutions but also in sterile and nonsterile wastewater.