Efficient p-value evaluation for resampling-based tests

The resampling-based test, which often relies on permutation or bootstrap procedures, has been widely used for statistical hypothesis testing when the asymptotic distribution of the test statistic is unavailable or unreliable. It requires repeated calculations of the test statistic on a large number of simulated data sets for its significance level assessment, and thus it could become very computationally intensive. Here, we propose an efficient p-value evaluation procedure by adapting the stochastic approximation Markov chain Monte Carlo algorithm. The new procedure can be used easily for estimating the p-value for any resampling-based test. We show through numeric simulations that the proposed procedure can be 100-500 000 times as efficient (in term of computing time) as the standard resampling-based procedure when evaluating a test statistic with a small p-value (e.g. less than 10( - 6)). With its computational burden reduced by this proposed procedure, the versatile resampling-based test would become computationally feasible for a much wider range of applications. We demonstrate the application of the new method by applying it to a large-scale genetic association study of prostate cancer.     

Large-scale bi-level strain design approaches and mixed-integer programming solution techniques

The use of computational models in metabolic engineering has been increasing as more genome-scale metabolic models and computational approaches become available. Various computational approaches have been developed to predict how genetic perturbations affect metabolic behavior at a systems level, and have been successfully used to engineer microbial strains with improved primary or secondary metabolite production. However, identification of metabolic engineering strategies involving a large number of perturbations is currently limited by computational resources due to the size of genome-scale models and the combinatorial nature of the problem. In this study, we present (i) two new bi-level strain design approaches using mixed-integer programming (MIP), and (ii) general solution techniques that improve the performance of MIP-based bi-level approaches. The first approach (SimOptStrain) simultaneously considers gene deletion and non-native reaction addition, while the second approach (BiMOMA) uses minimization of metabolic adjustment to predict knockout behavior in a MIP-based bi-level problem for the first time. Our general MIP solution techniques significantly reduced the CPU times needed to find optimal strategies when applied to an existing strain design approach (OptORF) (e.g., from ～10 days to ～5 minutes for metabolic engineering strategies with 4 gene deletions), and identified strategies for producing compounds where previous studies could not (e.g., malate and serine). Additionally, we found novel strategies using SimOptStrain with higher predicted production levels (for succinate and glycerol) than could have been found using an existing approach that considers network additions and deletions in sequential steps rather than simultaneously. Finally, using BiMOMA we found novel strategies involving large numbers of modifications (for pyruvate and glutamate), which sequential search and genetic algorithms were unable to find. The approaches and solution techniques developed here will facilitate the strain design process and extend the scope of its application to metabolic engineering.     

Interfacing interpreted and compiled languages to support applications on a massively parallel network of workstations (MP‐NOW)

The advent of Massively Parallel Network of Workstations (MP-NOW) represents an important trend in high performance computing. The rise of interpreted languages (e.g., Visual Basic, MATLAB, IDL, Maple and Mathematica) for algorithm development, prototyping, data analysis and graphical user interfaces (GUIs) represents an important trend in software engineering. However, using interpreted languages on a MP-NOW is a significant challenge. We present a specific example of a very simple, but generic solution to this problem. Our example uses an interpreted language to set up a calculation and then interfaces with a computational kernel written in a compiled language (e.g., C, C++, Fortran). The interpreted language calls the computational kernel as an external library. We have added to the computational kernel an additional layer, which manages multiple copies of the kernel running on a MP-NOW and returns the results back to the interpreted layer. Our implementation uses The Next generation Taskbag (TNT) library developed at Sarnoff to provide an efficient means for implementing task parallelism. A test problem (taken from Astronomy) has been implemented on the Sarnoff Cyclone computer which consists of 160 heterogeneous nodes connected by a “fat” tree 100 Mb/s switched Ethernet running the RedHat Linux and FreeBSD operating systems. Our first results in this ongoing project have demonstrated the feasibility of this approach and produced speedups of greater than 50 on 60 processors. This revised version was published online in July 2006 with corrections to the Cover Date.     

Improvement of the Teflonized Slides Used in the Immunofluoresgent Antibody Technique

The indirect fluorescent antibody (IF) test is often performed on glass slides prepared with smears or tissue sections containing an antigen. It is convenient to have several separate and distinct areas of antigen on the same slide, e.g. to compare dilution steps, and to save time when producing the antigen slides. With this end in view a technique for preparing slides with wells in a thin film of teflon-like compound has been elaborated (Goldman 1968). This technique is useful when a great number of slides need to be prepared in advance, e.g. to test a substantial amount of material. It has, however, not been widely used im serology for Babesia, perhaps because any teflon remaining within the wells is stained with the fluorescent dye, thus creating a potentially disturbing background.     

Graphle: Interactive exploration of large, dense graphs

Background  

A wide variety of biological data can be modeled as network structures, including experimental results (e.g. protein-protein interactions), computational predictions (e.g. functional interaction networks), or curated structures (e.g. the Gene Ontology). While several tools exist for visualizing large graphs at a global level or small graphs in detail, previous systems have generally not allowed interactive analysis of dense networks containing thousands of vertices at a level of detail useful for biologists. Investigators often wish to explore specific portions of such networks from a detailed, gene-specific perspective, and balancing this requirement with the networks' large size, complex structure, and rich metadata is a substantial computational challenge.     

DnaSAM: Software to perform neutrality testing for large datasets with complex null models

Patterns of DNA sequence polymorphisms can be used to understand the processes of demography and adaptation within natural populations. High-throughput generation of DNA sequence data has historically been the bottleneck with respect to data processing and experimental inference. Advances in marker technologies have largely solved this problem. Currently, the limiting step is computational, with most molecular population genetic software allowing a gene-by-gene analysis through a graphical user interface. An easy-to-use analysis program that allows both high-throughput processing of multiple sequence alignments along with the flexibility to simulate data under complex demographic scenarios is currently lacking. We introduce a new program, named DnaSAM, which allows high-throughput estimation of DNA sequence diversity and neutrality statistics from experimental data along with the ability to test those statistics via Monte Carlo coalescent simulations. These simulations are conducted using the ms program, which is able to incorporate several genetic parameters (e.g. recombination) and demographic scenarios (e.g. population bottlenecks). The output is a set of diversity and neutrality statistics with associated probability values under a user-specified null model that are stored in easy to manipulate text file.     

An overview of bioactivation of chemical carcinogens

Most environmental carcinogens require metabolic activation to reactive intermediates and are mutagenic in appropriate test systems. During the last decade, the cDNAs of numerous xenobiotic-metabolizing enzymes have been cloned. The individually expressed enzymes were used to study their substrate specificities and their inhibition by other compounds. Various enzymes were expressed directly in target cells of in vitro mutagenicity tests. This is illustrated in the present study for rat and human sulphotransferases (SULTs) expressed in Salmonella typhimurium TA1538. Numerous compounds were mutagenic in the new test system. Some of these promutagens were activated by several different SULT forms, whereas many other promutagens were activated with high selectivity by a specific enzyme form, but not by genetically closely related forms from the same species (e.g. allelic variants) or orthologous enzymes from other species. Similar findings have been made using recombinant test systems for specific forms of other classes of enzymes (e.g. cytochromes P450). This high selectivity in activation (and inactivation) may explain some organotropisms as well as species and inter-individual differences in the action of carcinogens. Many carcinogen-metabolizing enzymes are induced or inhibited by other xenobiotics. Such interactions can be exploited for chemo-prevention, which however may be carcinogen- and tissue-dependent.     

Niche conservatism as an emerging principle in ecology and conservation biology

The diversity of life is ultimately generated by evolution, and much attention has focused on the rapid evolution of ecological traits. Yet, the tendency for many ecological traits to instead remain similar over time [niche conservatism (NC)] has many consequences for the fundamental patterns and processes studied in ecology and conservation biology. Here, we describe the mounting evidence for the importance of NC to major topics in ecology (e.g. species richness, ecosystem function) and conservation (e.g. climate change, invasive species). We also review other areas where it may be important but has generally been overlooked, in both ecology (e.g. food webs, disease ecology, mutualistic interactions) and conservation (e.g. habitat modification). We summarize methods for testing for NC, and suggest that a commonly used and advocated method (involving a test for phylogenetic signal) is potentially problematic, and describe alternative approaches. We suggest that considering NC: (1) focuses attention on the within‐species processes that cause traits to be conserved over time, (2) emphasizes connections between questions and research areas that are not obviously related (e.g. invasives, global warming, tropical richness), and (3) suggests new areas for research (e.g. why are some clades largely nocturnal? why do related species share diseases?).     

Integrated knowledge mining,genome-scale modeling,and machine learning for predicting Yarrowia lipolytica bioproduction

Predicting bioproduction titers from microbial hosts has been challenging due to complex interactions between microbial regulatory networks, stress responses, and suboptimal cultivation conditions. This study integrated knowledge mining, feature extraction, genome-scale modeling (GSM), and machine learning (ML) to develop a model for predicting Yarrowia lipolytica chemical titers (i.e., organic acids, terpenoids, etc.). First, Y. lipolytica production data, including cultivation conditions, genetic engineering strategies, and product information, was manually collected from literature (~100 papers) and stored as either numerical (e.g., substrate concentrations) or categorical (e.g., bioreactor modes) variables. For each case recorded, central pathway fluxes were estimated using GSMs and flux balance analysis (FBA) to provide metabolic features. Second, a ML ensemble learner was trained to predict strain production titers. Accurate predictions on the test data were obtained for instances with production titers >1 g/L (R² = 0.87). However, the model had reduced predictability for low performance strains (0.01–1 g/L, R² = 0.29) potentially due to biosynthesis bottlenecks not captured in the features. Feature ranking indicated that the FBA fluxes, the number of enzyme steps, the substrate inputs, and thermodynamic barriers (i.e., Gibbs free energy of reaction) were the most influential factors. Third, the model was evaluated on other oleaginous yeasts and indicated there were conserved features for some hosts that can be potentially exploited by transfer learning. The platform was also designed to assist computational strain design tools (such as OptKnock) to screen genetic targets for improved microbial production in light of experimental conditions.     

Ethylene thiourea (ETU). A review of the genetic toxicity studies

Ethylene thiourea (ETU) is a common contaminant, metabolite and degradation product of the fungicide class of ethylene bisdithiocarbamates (EBDCs); as such, they present possible exposure and toxicological concerns to exposed individuals. ETU has been assayed in many different tests to assess genotoxicity activity. While a great number of negative results are found in the data base, there is evidence that demonstrates ETU is capable of inducing genotoxic endpoints. These include responses for gene mutations (e.g. Salmonella), structural chromosomal alterations (e.g. aberrations in cultured mammalian cells as well as a dominant lethal assay) and other genotoxic effects (e.g. bacterial rec assay and several yeast assays).It is important to consider the magnitude of the positive responses as well as the concentrations/doses used when assessing the genotoxicity of ETU. While ETU induces a variety of genotoxic endpoints, it does not appear to be a potent genotoxic agent. For example, it is a weak bacterial mutagen in the Salmonella assay without activation in strain TA1535 at concentrations generally above 1000 μg/plate. Weak genotoxic activity of this sort is usually observed in most of the assays with positive results. Since ETU does not appear very potent and is not extremely toxic to test cells and organisms, it is not surprising to find that ETU does not produce consistent effects in many of the assays reviewed. Consequently, in many instances, mixed results for the same assay type are reported by different investigators, but as reviewed herein, these results may be dependent upon the test conditions in each individual laboratory. A primary shortcoming with many of the reported negative results is that the concentrations or doses used are not high enough for an adequate test for ETU activity. There are also problems with many of the negative assays generally in protocol or reporting, particularly with the in vivo studies (e.g. inappropriate sample number and/or sampling times; inadequate top dose employed).Overall, while ETU does not appear to be a potent genotoxic agent, it is capable of producing genotoxic effects (e.g. gene mutations, structural chromosomal aberrations). This provides a basis for weak genotoxic activity by ETU. Furthermore, based on a suggestive dominant lethal positive result, there may be a concern for heritable effects. Due to the many problems with the conduct and assessment of the in vivo assays, it is worth repeating in vivo     

Microenvironment driven invasion: a multiscale multimodel investigation

Cancer is a complex, multiscale process, in which genetic mutations occurring at a subcellular level manifest themselves as functional and morphological changes at the cellular and tissue scale. The importance of interactions between tumour cells and their microenvironment is currently of great interest in experimental as well as computational modelling. Both the immediate microenvironment (e.g. cell-cell signalling or cell-matrix interactions) and the extended microenvironment (e.g. nutrient supply or a host tissue structure) are thought to play crucial roles in both tumour progression and suppression. In this paper we focus on tumour invasion, as defined by the emergence of a fingering morphology, which has previously been shown to be dependent upon harsh microenvironmental conditions. Using three different modelling approaches at two different spatial scales we examine the impact of nutrient availability as a driving force for invasion. Specifically we investigate how cell metabolism (the intrinsic rate of nutrient consumption and cell resistance to starvation) influences the growing tumour. We also discuss how dynamical changes in genetic makeup and morphological characteristics, of the tumour population, are driven by extreme changes in nutrient supply during tumour development. The simulation results indicate that aggressive phenotypes produce tumour fingering in poor nutrient, but not rich, microenvironments. The implication of these results is that an invasive outcome appears to be co-dependent upon the evolutionary dynamics of the tumour population driven by the microenvironment.     

A more efficient search strategy for aging genes based on connectivity

MOTIVATION: Many aging genes have been found from unbiased screens in model organisms. Genetic interventions promoting longevity are usually quantitative, while in many other biological fields (e.g. development) null mutations alone have been very informative. Therefore, in the case of aging the task is larger and the need for a more efficient genetic search strategy is especially strong. RESULTS: The topology of genetic and metabolic networks is organized according to a scale-free distribution, in which hubs with large numbers of links are present. We have developed a computational model of aging genes as the hubs of biological networks. The computational model shows that, after generalized damage, the function of a network with scale-free topology can be significantly restored by a limited intervention on the hubs. Analyses of data on aging genes and biological networks support the applicability of the model to biological aging. The model also might explain several of the properties of aging genes, including the high degree of conservation across different species. The model suggests that aging genes tend to have a higher number of connections and therefore supports a strategy, based on connectivity, for prioritizing what might otherwise be a random search for aging genes.     

Modeling catalytic reaction mechanisms in glycoside hydrolases

Modeling catalysis in carbohydrate-active enzymes is a daunting challenge because of the high flexibility and diversity of both enzymes and carbohydrates. Glycoside hydrolases (GHs) are an illustrative example, where conformational changes and subtle interactions have been shown to be critical for catalysis. GHs have pivotal roles in industry (e.g. biofuel or detergent production) and biomedicine (e.g. targets for cancer and diabetes), and thus, a huge effort is devoted to unveil their molecular mechanisms. Besides experimental techniques, computational methods have served to provide an in-depth understanding of GH mechanisms, capturing complex reaction coordinates and the conformational itineraries that substrates follow during the whole catalytic pathway, providing a framework that ultimately may assist the engineering of these enzymes and the design of new inhibitors.     

Cue integration in categorical tasks: insights from audio-visual speech perception

Previous cue integration studies have examined continuous perceptual dimensions (e.g., size) and have shown that human cue integration is well described by a normative model in which cues are weighted in proportion to their sensory reliability, as estimated from single-cue performance. However, this normative model may not be applicable to categorical perceptual dimensions (e.g., phonemes). In tasks defined over categorical perceptual dimensions, optimal cue weights should depend not only on the sensory variance affecting the perception of each cue but also on the environmental variance inherent in each task-relevant category. Here, we present a computational and experimental investigation of cue integration in a categorical audio-visual (articulatory) speech perception task. Our results show that human performance during audio-visual phonemic labeling is qualitatively consistent with the behavior of a Bayes-optimal observer. Specifically, we show that the participants in our task are sensitive, on a trial-by-trial basis, to the sensory uncertainty associated with the auditory and visual cues, during phonemic categorization. In addition, we show that while sensory uncertainty is a significant factor in determining cue weights, it is not the only one and participants' performance is consistent with an optimal model in which environmental, within category variability also plays a role in determining cue weights. Furthermore, we show that in our task, the sensory variability affecting the visual modality during cue-combination is not well estimated from single-cue performance, but can be estimated from multi-cue performance. The findings and computational principles described here represent a principled first step towards characterizing the mechanisms underlying human cue integration in categorical tasks.     

Meta-analyses of human gut microbes associated with obesity and IBD

Recent studies have linked human gut microbes to obesity and inflammatory bowel disease, but consistent signals have been difficult to identify. Here we test for indicator taxa and general features of the microbiota that are generally consistent across studies of obesity and of IBD, focusing on studies involving high-throughput sequencing of the 16S rRNA gene (which we could process using a common computational pipeline). We find that IBD has a consistent signature across studies and allows high classification accuracy of IBD from non-IBD subjects, but that although subjects can be classified as lean or obese within each individual study with statistically significant accuracy, consistent with the ability of the microbiota to experimentally transfer this phenotype, signatures of obesity are not consistent between studies even when the data are analyzed with consistent methods. The results suggest that correlations between microbes and clinical conditions with different effect sizes (e.g. the large effect size of IBD versus the small effect size of obesity) may require different cohort selection and analysis strategies.     

Modelling of non-linear elastic tissues for surgical simulation

Realistic modelling of the interaction between surgical instruments and human organs has been recognised as a key requirement in the development of high-fidelity surgical simulators. Primarily due to computational considerations, most of the past real-time surgical simulation research has assumed linear elastic behaviour for modelling tissues, even though human soft tissues generally possess non-linear properties. For a non-linear model, the well-known Poynting effect developed during shearing of the tissue results in normal forces not seen in a linear elastic model. Using constitutive equations of non-linear tissue models together with experiments, we show that the Poynting effect results in differences in force magnitude larger than the absolute human perception threshold for force discrimination in some tissues (e.g. myocardial tissues) but not in others (e.g. brain tissue simulants).     

A note on exact tests of Hardy-Weinberg equilibrium

Deviations from Hardy-Weinberg equilibrium (HWE) can indicate inbreeding, population stratification, and even problems in genotyping. In samples of affected individuals, these deviations can also provide evidence for association. Tests of HWE are commonly performed using a simple chi2 goodness-of-fit test. We show that this chi2 test can have inflated type I error rates, even in relatively large samples (e.g., samples of 1,000 individuals that include approximately 100 copies of the minor allele). On the basis of previous work, we describe exact tests of HWE together with efficient computational methods for their implementation. Our methods adequately control type I error in large and small samples and are computationally efficient. They have been implemented in freely available code that will be useful for quality assessment of genotype data and for the detection of genetic association or population stratification in very large data sets.     

Nonparametric Evaluations of Familial Aggregation

Limitations of detection of familial aggregation, examined by nonparametric aggregate measures (e.g., SMR), or by parametric methods (e.g., segregation analysis) may involve: (1) heterogeneity of risk patterns in different kindreds, or (2) restricted choice of parametric models. Nonparametric approaches dealing with individual kindreds, on the contrary, involve sparce data for which large sample theory may not apply. When the expected risk schedules of individuals are known, some optimal test procedures may be developed. Nevertheless, the optimality criteria are shown to be satisfied for only restricted choice of alternatives, and large sample approximation are not warranted in analysis of individual kindreds. Truncations of points of entry and exit of individuals in the study are needed to avoid complications due to loss of followup, in presence of which expected risk calculations are not reliable. Covariances of components of a goodness-of-fit test criterion detect specific modes of aggregation. It is illustrated that this test statistic is not always less powerful than the aggregate measures in detecting aggregation, and the covariances of its components partitioned by relative classes are in conformity with the results of segregation analysis.     

Current trends in mapping human genes

The human is estimated to have at least 50,000 expressed genes (gene loci). Some information is available concerning about 5000 of these gene loci and about 1900 have been mapped, i.e., assigned to specific chromosomes (and in most instances particular chromosome regions). Progress has been achieved by a combination of physical mapping (e.g., study of somatic cell hybrids and chromosomal in situ hybridization) and genetic mapping (e.g., genetic linkage studies). New methods for both physical and genetic mapping are expanding the armamentarium. The usefulness of the mapping information is already evident; the spin-off from the Human Genome Project (HGP) begins immediately. The complete nucleotide sequence is the ultimate map of the human genome. Sequencing, although already under way for limited segments of the genome, will await further progress in gene mapping, and in particular creation of contig maps for each chromosome. Meanwhile the technology of sequencing and sequence information handling will be developed. It is argued that the HGP is a new form of coordinated, interdisciplinary science; that its primary objective must be seen as the creation of a tool for biomedical research--a source book that will be the basis of study of variation and function for a long time; that the impact on scientist training will be salutary by relieving graduate students of useless drudgery and by training scientists competent in both molecular genetics and computational science; and that the funding of the HGP will have an insignificant negative effect on science funding generally, and indeed may have a beneficial effect through economy of scale and a focusing of attention on the excitement of biology and medical science.