Chemical carcinogenicity is an important safety issue for the evaluation of drugs and environmental pollutants. The Ames test is useful for detecting genotoxic hepatocarcinogens. However, the assessment of Ames-negative hepatocarcinogens depends on 2-year rodent bioassays. Alternative methods are desirable for the efficient identification of Ames-negative hepatocarcinogens. This study proposed a decision tree-based method using chemical-chemical interaction information for predicting hepatocarcinogens. It performs much better than that using molecular descriptors with accuracies of 86% and 76% for validation and independent test, respectively. Four important interacting chemicals with interpretable decision rules were identified and analyzed. With the high prediction performances, the acquired decision rules based on chemical-chemical interactions provide a useful prediction method and better understanding of Ames-negative hepatocarcinogens.
Microarray represents a recent multidisciplinary technology. It measures the expression levels of several genes under different biological conditions, which allows to generate multiple data. These data can be analyzed through biclustering method to determinate groups of genes presenting a similar behavior under specific groups of conditions. This paper proposes a new evolutionary algorithm based on a new crossover method, dedicated to the biclustering of gene expression data. This proposed crossover method ensures the creation of new biclusters with better quality. To evaluate its performance, an experimental study was done on real microarray datasets. These experimentations show that our algorithm extracts high quality biclusters with highly correlated genes that are particularly involved in specific ontology structure.
In recent years, several methods for gene networks (GNs) inference from expression data have been developed. Also, models of data integration (as protein-protein and protein-DNA) are nowadays broadly used to face the problem of few amount of expression data. Moreover, it is well known that biological networks conserve some topological properties. The small-world topology is a common arrangement in nature found both in biological and non-biological phenomena. However, in general this information is not used by GNs inference methods. In this work we proposed a new GNs inference algorithm that combines topological features and expression data. The algorithm outperforms the approach that uses only expression data both in accuracy and measures of recovered network.
Membrane protein prediction is a significant classification problem, requiring the integration of data derived from different sources such as protein sequences, gene expression, protein interactions etc. A generalized probabilistic approach for combining different data sources via supervised selective kernel fusion was proposed in our previous papers. It includes, as particular cases, SVM, Lasso SVM, Elastic Net SVM and others. In this paper we apply a further instantiation of this approach, the Supervised Selective Support Kernel SVM and demonstrate that the proposed approach achieves the top-rank position among the selective kernel fusion variants on benchmark data for membrane protein prediction. The method differs from the previous approaches in that it naturally derives a subset of "support kernels" (analogous to support objects within SVMs), thereby allowing the memory-efficient exclusion of significant numbers of irrelevant kernel matrixes from a decision rule in a manner particularly suited to membrane protein prediction.
Reverse engineering of the Gene Regulatory Networks (GRNs) from high-throughput gene expression data is one of the most pressing challenges of computational biology. In this paper a method for parallelization of the Gene Regulatory Network inference algorithm, GENIE3, based on GPU by exploiting the compute unified device architecture (CUDA) programming model is designed and implemented. GENIE3 solves regulatory network prediction by developing tree based ensemble of Random forests. Our proposed method significantly improves the computational efficiency of GENIE3 by constructing the forest on the GPU in parallel. Our experiments on real and synthetic datasets show that, CUDA implementation outperforms sequential implementation by achieving a speed-up of 15 times (real data) and 14 to 18 times (synthetic data) respectively.
Most of the current practice of pattern matching tools is oriented towards finding efficient ways to compare sequences. This is useful but insufficient: as the knowledge and understanding of some functional or structural aspects of living systems improve, analysts in molecular biology progressively shift from mere classification tasks to modeling tasks. People need to be able to express global sequence architectures and check various hypotheses on the way their sequences are structured. It appears necessary to offer generic tools for this task, allowing to build more expressive models of biological sequence families, on the basis of their content and structure. This article introduces Logol, a new application designed to achieve pattern matching in possibly large sequences with customized biological patterns. Logol consists in both a language for describing patterns, and the associated parser for effective pattern search in sequences (RNA, DNA or protein) with such patterns. The Logol language, based on an high level grammatical formalism, allows to express flexible patterns (with mispairings and indels) composed of both sequential elements (such as motifs) and structural elements (such as repeats or pseudoknots). Its expressive power is presented through an application using the main components of the language : the identification of -1 programmed ribosomal frameshifting (PRF) events in messenger RNA sequences. Logol allows the design of sophisticated patterns, and their search in large nucleic or amino acid sequences. It is available on the GenOuest bioinformatics platform at http://logol.genouest.org. The core application is a command-line application, available for different operating systems. The Logol suite also includes interfaces, e.g. an interface for graphically drawing the pattern.
Network based prediction of interaction between drug compounds and target proteins is a core step in the drug discovery process. The availability of drug–target interaction data has boosted the development of machine learning methods for the in silico prediction of drug– target interactions. In this paper we focus on the crucial issue of data bias. We show that four popular datasets contain a bias because of the way they have been constructed: all drug compounds and target proteins have at least one interaction and some of them have only a single interaction. We show that this bias can be exploited by prediction methods to achieve an optimistic generalization performance as estimated by crossvalidation procedures, in particular leave-one-out cross validation. We discuss possible ways to mitigate the effect of this bias, in particular by adapting the validation procedure. In general, results indicate that the data bias should be taken into account when assessing the generalization performance of machine learning methods for the in silico prediction of drug–target interactions. The datasets and source code for this article are available at http://cs. ru.nl/~tvanlaarhoven/bias2014/
Microarrays represent a new technology for measuring expression levels of several genes under various biological conditions generating multiple data. These data can be analyzed by using biclustering method which aims to extract a maximum number of genes and conditions presenting a similar behavior. This paper proposes a new evolutionary approach to obtain maximal high-quality biclusters of highly-correlated genes. The performance of the proposed algorithm is assessed on synthetic gene expression data. Experimental results show that our algorithm competes favorably with several state-of-the-art biclustering algorithms.
In Cytomics, the study of cellular systems at the single cell level, High-Throughput Screening (HTS) techniques have been developed to implement the testing of hundreds to thousands of conditions applied to several or up to millions of cells in a single experiment.Recent technological developments of imaging systems and robotics have lead to an exponential increase in data volumes generated in HTS-experiments. This is pushing forward the need for a semantically oriented bioinformatics approach capable of storing large volume of linked metadata, handling a diversity of data formats, and querying data in order to extract meaning from the experiments performed.This paper describes our research in developing CytomicsDB, a modern RDBMS based platform, designed to provide an architecture capable of dealing with the computational requirements involved in high-throughput content. CytomicsDB supports web services and collaborative infrastructure in order to perform further exploration of linked information generated in each experiment.The objective of this system is to build a semantic layer over the data so as to enable querying metadata and at the same time allowing scientists to integrate new tools and APIs taking care of the image and data analysis. The results will become part of the metadata of the whole HTS experiment and will be available for semantic post analysis.
The reconstruction of protein-protein interaction networks is nowadays an important challenge in systems biology. Computational approaches can address this problem by complementing high-throughput technologies and by helping and guiding biologists in designing new laboratory experiments. The proteins and the interactions between them form a network, which has been shown to possess several topological properties. In addition to information about proteins and interactions between them, knowledge about the topological properties of these networks can be used to learn accurate models for predicting unknown protein-protein interactions. This paper presents a principled way, based on Bayesian inference, for combining network topology information jointly with information about proteins and interactions between them. The goal of this combination is to build accurate models for predicting protein-protein interactions. We define a random graph model for generating networks with topology similar to the ones observed in protein-protein interaction networks. We define a probability model for protein features given the absence/presence of an interaction and combine this with the random graph model by using Bayes’ rule, to finally arrive at a model incorporating both topological and feature information.
Determining the structural class of a given protein can provide important information about its functionality and its general tertiary structure. In the last two decades, the protein structural class prediction problem has attracted tremendous attention and its prediction accuracy has been significantly improved. Features extracted from the Position Specific Scoring Matrix (PSSM) have played an important role to achieve this enhancement. However, this information has not been adequately explored since the protein structural class prediction accuracy relying on PSSM for feature extraction still remains limited. In this study, to explore this potential, we propose segmentation-based feature extraction technique based on the concepts of amino acids' distribution and auto covariance. By applying a Support Vector Machine (SVM) to our extracted features, we enhance protein structural class prediction accuracy up to 16% over similar studies found in the literature. We achieve over 90% and 80% prediction accuracies for 25PDB and 1189 benchmarks respectively by solely relying on the PSSM for feature extraction.
T-score, based on t-statistics between samples and disease classes, is a widely used filter criterion for gene selection from microarray data. However, classical T-score uses all the training samples but for both biological and computational reasons, selection of relevant samples for training is an important step in classification. Using a modified logistic regression approach, we propose a sample selection criterion based on T-score and develop a backward elimination approach for gene selection. The method is more stable and computationally less costly compared to support vector machine recursive feature elimination (SVM-RFE) methods.
Markov Random Fields (MRF) have been shown to be good predictors of functional annotation, using protein-protein interaction data. Many other sources of data can also be used in this prediction task, but they are typically not integrated.In this study, we extend a method using MRFs in order to allow the use of additional data. A conditional random field (CRF) model is proposed as an alternative to an MRF model in order to remove the requirement of modeling relationships between the sources of data. We observe that a substantial performance improvement is possible using additional data, such as genetic interaction networks. The improvement gained from each source of evidence is not the same for each protein function, indicating that each source supplies different information. We demonstrate that CRFs can be used to efficiently integrate various sources of data to predict functional annotations.
The relationship connecting the biomolecular sequence, the molecular structure, and the biological function is of extreme importance in nanostructure analysis such as drug discovery. Previous studies involving multiple sequence alignment of biomolecules have demonstrated that associated sites are indicative of the structural and functional characteristics of biomolecules, comparable to methods such as consensus sequences analysis. In this paper, a new method to detect associated sites in aligned sequence ensembles is proposed. It involves the use of multiple sub-tables (or levels) of two-dimensional contingency table analysis. The idea is to incorporate analysis by using a concept known as granular computing, which represents information at different levels of granularity. The analysis involves two phases. The first phase includes labeling of the molecular sites in the p53 protein multiple sequence alignment according to the detected associated patterns. The sites are consequently labeled into three different types based on their site characteristics: 1) conserved sites, 2) associated sites and 3) hypervariate sites. In the second phase, the significance of the extracted site patterns is evaluated with respect to targeted structural and functional characteristics of the p53 protein. The results indicate that the extracted site patterns are significantly associated with some of the known functionalities of p53, a cancer suppressor. Furthermore, when these sites are aligned with p63 and p73, the homologs of p53 without the same cancer suppressing property, based on the common domains, the sites significantly discriminate between the human sequences of the p53 family. Therefore, the study confirms the importance of these detected sites that could indicate their differences in cancer suppressing property.
The effects of residue substitution in protein can be dramatic and predicting its impact may benefit scientists greatly. Like in many scientific domains there are various methods and tools available to address the potential impact of a mutation on the structure of a protein. The identification of these methods, their availability, the time needed to gain enough familiarity with them and their interface, and the difficulty of integrating their results in a global view where all view points can be visualized often limit their use. In this paper, we present the Structural Prediction for pRotein fOlding UTility System (SPROUTS) workflow and describe our method for designing, documenting, and maintaining the workflow. The focus of the workflow is the thermodynamic contribution to stability, which can be considered as acceptable for small proteins. It compiles the predictions from various sources calculating the ΔΔG upon point mutation, together with a consensus from eight distinct algorithms, with a prediction of the mean number of interacting residues during the process of folding, and a sub domain structural analysis into fragments that may potentially be considered as autonomous folding units, i.e., with similar conformations alone and in the protein body. The workflow is implemented and available online. We illustrate its use with the analysis of the engrailed homeodomain (PDB code 1enh).
Feature weighting in supervised learning concerns the development of methods for quantifying the capability of features to discriminate instances from different classes. A popular method for this task, called RELIEF, generates a feature weight vector from a given training set, one weight for each feature. This is achieved by maximizing in a greedy way the sample margin defined on the nearest neighbor classifier. The contribution from each class to the sample margin maximization defines a set of class dependent feature weight vectors, one for each class. This provides a tool to unravel interesting properties of features relevant to a single class of interest. In this paper we analyze such class dependent feature weight vectors. For instance, we show that in a machine learning dataset describing instances of recurrence and non-recurrence events in breast cancer, the features have different relevance in the two types of events, with size of the tumor estimated to be highly relevant in the recurrence class but not in the non-recurrence one. Furthermore, results of experiments show that a high correlation between feature weights of one class and those generated by RELIEF corresponds to an easier classification task. In general, results of this investigation indicate that class dependent feature weights are useful to unravel interesting properties of features with respect to a class of interest, and they provide information on the relative difficulty of classification tasks.
The high-throughput technologies have led to vast amounts of protein-protein interaction (PPI) data, and a number of approaches based on PPI networks have been proposed for protein function prediction. However, these approaches do not work well if annotated proteins are scarce in the networks. To address this issue, we propose an active learning based approach that uses graph-based centrality metrics to select proper candidates for labeling. We first cluster a PPI network by using the spectral clustering algorithm and select some proper candidates for labeling within each cluster, and then apply a collective classification algorithm to predict protein function based on these annotated proteins. Experiments over two real datasets demonstrate that the active learning based approach achieves better prediction performance by choosing more informative proteins for labeling. Experimental results also validate that betweenness centrality is more effective than degree centrality and closeness centrality in most cases.
Proteins are known to interact with each other by forming protein complexes and in order to perform specific biological functions. Many community detection methods have been devised for the discovery of protein complexes in protein interaction networks. One common problem in current agglomerative community detection approaches is that vertices with just one neighbor are often classified as separate clusters, which does not make sense for complex identification. Also, a major limitation of agglomerative techniques is that their computational efficiency do not scale well to large protein interaction networks (PINs). In this paper, we propose a new agglomerative algorithm, FAC-PIN, based on a local premetric of relative vertex-to-vertex clustering value and which addresses the above two issues. Our proposed FAC-PIN method is applied to eight PINs from different species, and the identified complexes are validated using experimentally verified complexes. The preliminary computational results show that FAC-PIN can discover protein complexes from PINs more accurately and faster than the HC-PIN and CNM algorithms, the current state-of-the-art agglomerative approaches to complex prediction.
Protein fold recognition (PFR) is considered as an important step towards the protein structure prediction problem. It also provides crucial information about the functionality of the proteins. Despite all the efforts that have been made during the past two decades, finding an accurate and fast computational approach to solve PFR still remains a challenging problem for bioinformatics and computational biology. It has been shown that extracting features which contain significant local and global discriminatory information plays a key role in addressing this problem. In this study, we propose the concept of segmented-based feature extraction technique to provide local evolutionary information embedded in Position Specific Scoring Matrix (PSSM) and structural information embedded in the predicted secondary structure of proteins using SPINE-X. We also employ the concept of occurrence feature to extract global discriminatory information from PSSM and SPINE-X. By applying a Support Vector Machine (SVM) to our extracted features, we enhance the protein fold prediction accuracy to 7.4% over the best results reported in the literature.
Viral haplotype estimation in a population is an important problem in virology. Viruses undergo a high number of mutations and recombinations during replication for their survival in host cells and exist as a population of closely related genetic variants. Due to this, estimating the number of haplotypes and their relative frequencies in the population becomes a challenging task. The usage of a sequenced reference genome has its limitations due to the high mutational rates in viruses. We propose a method for estimating viral haplotypes based only on the counts of k-mers present in the viral population without using the reference genome. We compute k-mer pairs that are related to each other by one mutation, and compute a minimal set of viral haplotypes that explain the whole population based on these k-mer pairs. We compare our method to the software ShoRAH (which uses a reference genome) on simulated dataset and obtained comparable results, even without using a reference genome.
