GANN: genetic algorithm neural networks for the detection of conserved combinations of features in DNA

Artificial neural networks predict the incidence of portosplenomesenteric venous thrombosis in patients with acute pancreatitis

Essentials Predicting the occurrence of portosplenomesenteric vein thrombosis (PSMVT) is difficult. We studied 72 patients with acute pancreatitis. Artificial neural networks modeling was more accurate than logistic regression in predicting PSMVT. Additional predictive factors may be incorporated into artificial neural networks.

SUMMARY

Objective To construct and validate artificial neural networks (ANNs) for predicting the occurrence of portosplenomesenteric venous thrombosis (PSMVT) and compare the predictive ability of the ANNs with that of logistic regression. Methods The ANNs and logistic regression modeling were constructed using simple clinical and laboratory data of 72 acute pancreatitis (AP) patients. The ANNs and logistic modeling were first trained on 48 randomly chosen patients and validated on the remaining 24 patients. The accuracy and the performance characteristics were compared between these two approaches by SPSS17.0 software. Results The training set and validation set did not differ on any of the 11 variables. After training, the back propagation network training error converged to 1 × 10-20 , and it retained excellent pattern recognition ability. When the ANNs model was applied to the validation set, it revealed a sensitivity of 80%, specificity of 85.7%, a positive predictive value of 77.6% and negative predictive value of 90.7%. The accuracy was 83.3%. Differences could be found between ANNs modeling and logistic regression modeling in these parameters (10.0% [95% CI, -14.3 to 34.3%], 14.3% [95% CI, -8.6 to 37.2%], 15.7% [95% CI, -9.9 to 41.3%], 11.8% [95% CI, -8.2 to 31.8%], 22.6% [95% CI, -1.9 to 47.1%], respectively). When ANNs modeling was used to identify PSMVT, the area under receiver operating characteristic curve was 0.849 (95% CI, 0.807-0.901), which demonstrated better overall properties than logistic regression modeling (AUC = 0.716) (95% CI, 0.679-0.761). Conclusions ANNs modeling was a more accurate tool than logistic regression in predicting the occurrence of PSMVT following AP. More clinical factors or biomarkers may be incorporated into ANNs modeling to improve its predictive ability.

YGA: identifying distinct biological features between yeast gene sets

The advance of high-throughput experimental technologies generates many gene sets with different biological meanings, where many important insights can only be extracted by identifying the biological (regulatory/functional) features that are distinct between different gene sets (e.g. essential vs. non-essential genes, TATA box-containing vs. TATA box-less genes, induced vs. repressed genes under certain biological conditions). Although many servers have been developed to identify enriched features in a gene set, most of them were designed to analyze one gene set at a time but cannot compare two gene sets. Moreover, the features used in existing servers were mainly focused on functional annotations (GO terms), pathways, transcription factor binding sites (TFBSs) and/or protein-protein interactions (PPIs). In yeast, various important regulatory features, including promoter bendability, nucleosome occupancy, 5'-UTR length, and TF-gene regulation evidence, are available but have not been used in any enrichment analysis servers. This motivates us to develop the Yeast Genes Analyzer (YGA), a web server that simultaneously analyzes various biological (regulatory/functional) features of two gene sets and performs statistical tests to identify the distinct features between them. Many well-studied gene sets such as essential, stress-response, TATA box-containing and cell cycle genes were pre-compiled in YGA for users, if they have only one gene set, to compare with. In comparison with the existing enrichment analysis servers, YGA tests more comprehensive regulatory features (e.g. promoter bendability, nucleosome occupancy, 5'-UTR length, experimental evidence of TF-gene binding and TF-gene regulation) and functional features (e.g. PPI, GO terms, pathways and functional groups of genes, including essential/non-essential genes, stress-induced/-repressed genes, TATA box-containing/-less genes, occupied/depleted proximal-nucleosome genes and cell cycle genes). Furthermore, YGA uses various statistical tests to provide objective comparison measures. The two major contributions of YGA, comprehensive features and statistical comparison, help to mine important information that cannot be obtained from other servers. The sophisticated analysis tools of YGA can identify distinct biological features between two gene sets, which help biologists to form new hypotheses about the underlying biological mechanisms responsible for the observed difference between these two gene sets. YGA can be accessed from the following web pages: http://cosbi.ee.ncku.edu.tw/yga/ and http://yga.ee.ncku.edu.tw/.

DNA structural properties in the classification of genomic transcription regulation elements

It has been long known that DNA molecules encode information at various levels. The most basic level comprises the base sequence itself and is primarily important for the encoding of proteins and direct base recognition by DNA-binding proteins. A more elusive level consists of the local structural properties of the DNA molecule wherein the DNA sequence only plays an indirect supportive role. These properties are nevertheless an important factor in a large number of biomolecular processes and can be considered as informative signals for the presence of a variety of genomic features. Several recent studies have unequivocally shown the benefit of relying on such DNA properties for modeling and predicting genomic features as diverse as transcription start sites, transcription factor binding sites, or nucleosome occupancy. This review is meant to provide an overview of the key aspects of these DNA conformational and physicochemical properties. To illustrate their potential added value compared to relying solely on the nucleotide sequence in genomics studies, we discuss their application in research on transcription regulation mechanisms as representative cases.

GRISOTTO: A greedy approach to improve combinatorial algorithms for motif discovery with prior knowledge

BACKGROUND

Position-specific priors (PSP) have been used with success to boost EM and Gibbs sampler-based motif discovery algorithms. PSP information has been computed from different sources, including orthologous conservation, DNA duplex stability, and nucleosome positioning. The use of prior information has not yet been used in the context of combinatorial algorithms. Moreover, priors have been used only independently, and the gain of combining priors from different sources has not yet been studied.

RESULTS

We extend RISOTTO, a combinatorial algorithm for motif discovery, by post-processing its output with a greedy procedure that uses prior information. PSP's from different sources are combined into a scoring criterion that guides the greedy search procedure. The resulting method, called GRISOTTO, was evaluated over 156 yeast TF ChIP-chip sequence-sets commonly used to benchmark prior-based motif discovery algorithms. Results show that GRISOTTO is at least as accurate as other twelve state-of-the-art approaches for the same task, even without combining priors. Furthermore, by considering combined priors, GRISOTTO is considerably more accurate than the state-of-the-art approaches for the same task. We also show that PSP's improve GRISOTTO ability to retrieve motifs from mouse ChiP-seq data, indicating that the proposed algorithm can be applied to data from a different technology and for a higher eukaryote.

CONCLUSIONS

The conclusions of this work are twofold. First, post-processing the output of combinatorial algorithms by incorporating prior information leads to a very efficient and effective motif discovery method. Second, combining priors from different sources is even more beneficial than considering them separately.

Using genetic algorithms to select most predictive protein features

Many important characteristics of proteins such as biochemical activity and subcellular localization present a challenge to machine-learning methods: it is often difficult to encode the appropriate input features at the residue level for the purpose of making a prediction for the entire protein. The problem is usually that the biophysics of the connection between a machine-learning method's input (sequence feature) and its output (observed phenomenon to be predicted) remains unknown; in other words, we may only know that a certain protein is an enzyme (output) without knowing which region may contain the active site residues (input). The goal then becomes to dissect a protein into a vast set of sequence-derived features and to correlate those features with the desired output. We introduce a framework that begins with a set of global sequence features and then vastly expands the feature space by generically encoding the coexistence of residue-based features. It is this combination of individual features, that is the step from the fractions of serine and buried (input space 20 + 2) to the fraction of buried serine (input space 20 * 2) that implicitly shifts the search space from global feature inputs to features that can capture very local evidence such as a the individual residues of a catalytic triad. The vast feature space created is explored by a genetic algorithm (GA) paired with neural networks and support vector machines. We find that the GA is critical for selecting combinations of features that are neither too general resulting in poor performance, nor too specific, leading to overtraining. The final framework manages to effectively sample a feature space that is far too large for exhaustive enumeration. We demonstrate the power of the concept by applying it to prediction of protein enzymatic activity.

Comparative analysis of regulatory motif discovery tools for transcription factor binding sites

In the post-genomic era, identification of specific regulatory motifs or transcription factor binding sites (TFBSs) in non-coding DNA sequences, which is essential to elucidate transcriptional regulatory networks, has emerged as an obstacle that frustrates many researchers. Consequently, numerous motif discovery tools and correlated databases have been applied to solving this problem. However, these existing methods, based on different computational algorithms, show diverse motif prediction efficiency in non-coding DNA sequences. Therefore, understanding the similarities and differences of computational algorithms and enriching the motif discovery literatures are important for users to choose the most appropriate one among the online available tools. Moreover, there still lacks credible criterion to assess motif discovery tools and instructions for researchers to choose the best according to their own projects. Thus integration of the related resources might be a good approach to improve accuracy of the application. Recent studies integrate regulatory motif discovery tools with experimental methods to offer a complementary approach for researchers, and also provide a much-needed model for current researches on transcriptional regulatory networks. Here we present a comparative analysis of regulatory motif discovery tools for TFBSs.

Predicting peptides binding to MHC class II molecules using multi-objective evolutionary algorithms

Background

Peptides binding to Major Histocompatibility Complex (MHC) class II molecules are crucial for initiation and regulation of immune responses. Predicting peptides that bind to a specific MHC molecule plays an important role in determining potential candidates for vaccines. The binding groove in class II MHC is open at both ends, allowing peptides longer than 9-mer to bind. Finding the consensus motif facilitating the binding of peptides to a MHC class II molecule is difficult because of different lengths of binding peptides and varying location of 9-mer binding core. The level of difficulty increases when the molecule is promiscuous and binds to a large number of low affinity peptides.

In this paper, we propose two approaches using multi-objective evolutionary algorithms (MOEA) for predicting peptides binding to MHC class II molecules. One uses the information from both binders and non-binders for self-discovery of motifs. The other, in addition, uses information from experimentally determined motifs for guided-discovery of motifs.

Results

The proposed methods are intended for finding peptides binding to MHC class II I-A^g7 molecule – a promiscuous binder to a large number of low affinity peptides. Cross-validation results across experiments on two motifs derived for I-A^g7 datasets demonstrate better generalization abilities and accuracies of the present method over earlier approaches. Further, the proposed method was validated and compared on two publicly available benchmark datasets: (1) an ensemble of qualitative HLA-DRB1*0401 peptide data obtained from five different sources, and (2) quantitative peptide data obtained for sixteen different alleles comprising of three mouse alleles and thirteen HLA alleles. The proposed method outperformed earlier methods on most datasets, indicating that it is well suited for finding peptides binding to MHC class II molecules.

Conclusion

We present two MOEA-based algorithms for finding motifs, one for self-discovery and the other for guided-discovery by experimentally determined motifs, and thereby predicting binding peptides to I-A^g7 molecule. Our experiments show that the proposed MOEA-based algorithms are better than earlier methods in predicting binding sites not only on I-A^g7 but also on most alleles of class II MHC benchmark datasets. This shows that our methods could be applicable to find binding motifs in a wide range of alleles.

Complexity of type 2 diabetes mellitus data sets emerging from nutrigenomic research: a case for dimensionality reduction?

Nutrigenomics promises personalized nutrition and an improvement in preventing, delaying, and reducing the symptoms of chronic diseases such as diabetes. Nutritional genomics is the study of how foods affect the expression of genetic information in an individual and how an individual's genetic makeup affects the metabolism and response to nutrients and other bioactive components in food. The path to those promises has significant challenges, from experimental designs that include analysis of genetic heterogeneity to the complexities of food and environmental factors. One of the more significant complications in developing the knowledge base and potential applications is how to analyze high-dimensional datasets of genetic, nutrient, metabolomic (clinical), and other variables influencing health and disease processes. Type 2 diabetes mellitus (T2DM) is used as an illustration of the challenges in studying complex phenotypes with nutrigenomics concepts and approaches.

Virulence profiling of Streptococcus dysgalactiae subspecies equisimilis isolated from infected humans reveals 2 distinct genetic lineages that do not segregate with their phenotypes or propensity to cause diseases

BACKGROUND

In spite of the emerging importance of Streptococcus dysgalactiae subspecies equisimilis (human group C streptococci [GCS] and group G streptococci [GGS]) in human health, its molecular makeup remains largely undefined. Apart from sharing a phylogenetic relationship with the human pathogen group A streptococci (GAS), GCS/GGS and GAS colonize the same ecological niche and exhibit considerable overlap in their disease profiles. Such similarities imply that the virulence factors associated with diseases may also be similar.

METHODS

In this study, we used a targeted microarray containing 216 GAS virulence genes to profile the virulence gene repertoires of 58 S. dysgalactiae subspecies equisimilis isolates recovered during human infections. We performed comparative analyses to investigate the relationship between GAS virulence genes in and the invasive potential of GCS/GGS.

RESULTS

Up to one-half of the GAS virulence genes represented in the microarray were identified in GCS/GGS. No statistical differences were observed between isolates harboring the group C versus group G carbohydrates; however, clustering algorithms revealed 2 genetically distinct clusters of S. dysgalactiae subspecies equisimilis isolates. No relationship was observed between the virulence profile of GCS/GGS and the propensity for disease or the tissue site of isolation.

CONCLUSIONS

This is, to our knowledge, the first comprehensive analysis of the virulence profile of S. dysgalactiae subspecies equisimilis, and it enables novel insights into the pathogen's genetic basis of disease propensity shared with GAS. Human group C and group G streptococci may not be considered to be separate species; in fact, they may constitute 2 distinct lineages. Additional incongruent relationships were observed between virulence profiles and GCS/GGS disease propensity.

Recent computational approaches to understand gene regulation: mining gene regulation in silico

This paper reviews recent computational approaches to the understanding of gene regulation in eukaryotes. Cis-regulation of gene expression by the binding of transcription factors is a critical component of cellular physiology. In eukaryotes, a number of transcription factors often work together in a combinatorial fashion to enable cells to respond to a wide spectrum of environmental and developmental signals. Integration of genome sequences and/or Chromatin Immunoprecipitation on chip data with gene-expression data has facilitated in silico discovery of how the combinatorics and positioning of transcription factors binding sites underlie gene activation in a variety of cellular processes.The process of gene regulation is extremely complex and intriguing, therefore all possible points of view and related links should be carefully considered. Here we attempt to collect an inventory, not claiming it to be comprehensive and complete, of related computational biological topics covering gene regulation, which may en-lighten the process, and briefly review what is currently occurring in these areas.We will consider the following computational areas:o gene regulatory network construction;o evolution of regulatory DNA;o studies of its structural and statistical informational properties;o and finally, regulatory RNA.

Genes for the majority of group a streptococcal virulence factors and extracellular surface proteins do not confer an increased propensity to cause invasive disease

BACKGROUND

The factors behind the reemergence of severe, invasive group A streptococcal (GAS) diseases are unclear, but it could be caused by altered genetic endowment in these organisms. However, data from previous studies assessing the association between single genetic factors and invasive disease are often conflicting, suggesting that other, as-yet unidentified factors are necessary for the development of this class of disease.

METHODS

In this study, we used a targeted GAS virulence microarray containing 226 GAS genes to determine the virulence gene repertoires of 68 GAS isolates (42 associated with invasive disease and 28 associated with noninvasive disease) collected in a defined geographic location during a contiguous time period. We then employed 3 advanced machine learning methods (genetic algorithm neural network, support vector machines, and classification trees) to identify genes with an increased association with invasive disease.

RESULTS

Virulence gene profiles of individual GAS isolates varied extensively among these geographically and temporally related strains. Using genetic algorithm neural network analysis, we identified 3 genes with a marginal overrepresentation in invasive disease isolates. Significantly, 2 of these genes, ssa and mf4, encoded superantigens but were only present in a restricted set of GAS M-types. The third gene, spa, was found in variable distributions in all M-types in the study.

CONCLUSIONS

Our comprehensive analysis of GAS virulence profiles provides strong evidence for the incongruent relationships among any of the 226 genes represented on the array and the overall propensity of GAS to cause invasive disease, underscoring the pathogenic complexity of these diseases, as well as the importance of multiple bacteria and/or host factors.

A survey of motif discovery methods in an integrated framework

BACKGROUND

There has been a growing interest in computational discovery of regulatory elements, and a multitude of motif discovery methods have been proposed. Computational motif discovery has been used with some success in simple organisms like yeast. However, as we move to higher organisms with more complex genomes, more sensitive methods are needed. Several recent methods try to integrate additional sources of information, including microarray experiments (gene expression and ChlP-chip). There is also a growing awareness that regulatory elements work in combination, and that this combinatorial behavior must be modeled for successful motif discovery. However, the multitude of methods and approaches makes it difficult to get a good understanding of the current status of the field.

RESULTS

This paper presents a survey of methods for motif discovery in DNA, based on a structured and well defined framework that integrates all relevant elements. Existing methods are discussed according to this framework.

CONCLUSION

The survey shows that although no single method takes all relevant elements into consideration, a very large number of different models treating the various elements separately have been tried. Very often the choices that have been made are not explicitly stated, making it difficult to compare different implementations. Also, the tests that have been used are often not comparable. Therefore, a stringent framework and improved test methods are needed to evaluate the different approaches in order to conclude which ones are most promising.

REVIEWERS

This article was reviewed by Eugene V. Koonin, Philipp Bucher (nominated by Mikhail Gelfand) and Frank Eisenhaber.