An Information-Theoretic Machine Learning Approach to Expression QTL Analysis

Expression Quantitative Trait Locus (eQTL) analysis is a powerful tool to study the biological mechanisms linking the genotype with gene expression. Such analyses can identify genomic locations where genotypic variants influence the expression of genes, both in close proximity to the variant (cis-eQTL), and on other chromosomes (trans-eQTL). Many traditional eQTL methods are based on a linear regression model. In this study, we propose a novel method by which to identify eQTL associations with information theory and machine learning approaches. Mutual Information (MI) is used to describe the association between genetic marker and gene expression. MI can detect both linear and non-linear associations. What’s more, it can capture the heterogeneity of the population. Advanced feature selection methods, Maximum Relevance Minimum Redundancy (mRMR) and Incremental Feature Selection (IFS), were applied to optimize the selection of the affected genes by the genetic marker. When we applied our method to a study of apoE-deficient mice, it was found that the cis-acting eQTLs are stronger than trans-acting eQTLs but there are more trans-acting eQTLs than cis-acting eQTLs. We compared our results (mRMR.eQTL) with R/qtl, and MatrixEQTL (modelLINEAR and modelANOVA). In female mice, 67.9% of mRMR.eQTL results can be confirmed by at least two other methods while only 14.4% of R/qtl result can be confirmed by at least two other methods. In male mice, 74.1% of mRMR.eQTL results can be confirmed by at least two other methods while only 18.2% of R/qtl result can be confirmed by at least two other methods. Our methods provide a new way to identify the association between genetic markers and gene expression. Our software is available from supporting information.     

Understanding Confounding Effects in Linguistic Coordination: An Information-Theoretic Approach

We suggest an information-theoretic approach for measuring stylistic coordination in dialogues. The proposed measure has a simple predictive interpretation and can account for various confounding factors through proper conditioning. We revisit some of the previous studies that reported strong signatures of stylistic accommodation, and find that a significant part of the observed coordination can be attributed to a simple confounding effect—length coordination. Specifically, longer utterances tend to be followed by longer responses, which gives rise to spurious correlations in the other stylistic features. We propose a test to distinguish correlations in length due to contextual factors (topic of conversation, user verbosity, etc.) and turn-by-turn coordination. We also suggest a test to identify whether stylistic coordination persists even after accounting for length coordination and contextual factors.     

Information-Theoretic Analysis of the Dynamics of an Executable Biological Model

To facilitate analysis and understanding of biological systems, large-scale data are often integrated into models using a variety of mathematical and computational approaches. Such models describe the dynamics of the biological system and can be used to study the changes in the state of the system over time. For many model classes, such as discrete or continuous dynamical systems, there exist appropriate frameworks and tools for analyzing system dynamics. However, the heterogeneous information that encodes and bridges molecular and cellular dynamics, inherent to fine-grained molecular simulation models, presents significant challenges to the study of system dynamics. In this paper, we present an algorithmic information theory based approach for the analysis and interpretation of the dynamics of such executable models of biological systems. We apply a normalized compression distance (NCD) analysis to the state representations of a model that simulates the immune decision making and immune cell behavior. We show that this analysis successfully captures the essential information in the dynamics of the system, which results from a variety of events including proliferation, differentiation, or perturbations such as gene knock-outs. We demonstrate that this approach can be used for the analysis of executable models, regardless of the modeling framework, and for making experimentally quantifiable predictions.     

Pyrosequencing Characterization of the Microbiota from Atlantic Intertidal Marine Sponges Reveals High Microbial Diversity and the Lack of Co-Occurrence Patterns

Sponges are ancient metazoans that host diverse and complex microbial communities. Sponge-associated microbial diversity has been studied from wide oceans across the globe, particularly in subtidal regions, but the microbial communities from intertidal sponges have remained mostly unexplored. Here we used pyrosequencing to characterize the microbial communities in 12 different co-occurring intertidal marine sponge species sampled from the Atlantic coast, revealing a total of 686 operational taxonomic units (OTUs) at 97% sequence similarity. Taxonomic assignment of 16S ribosomal RNA tag sequences estimated altogether 26 microbial groups, represented by bacterial (75.5%) and archaeal (22%) domains. Proteobacteria (43.4%) and Crenarchaeota (20.6%) were the most dominant microbial groups detected in all the 12 marine sponge species and ambient seawater. The Crenarchaeota microbes detected in three Atlantic Ocean sponges had a close similarity with Crenarchaeota from geographically separated subtidal Red Sea sponges. Our study showed that most of the microbial communities observed in sponges (73%) were also found in the surrounding ambient seawater suggesting possible environmental acquisition and/or horizontal transfer of microbes. Beyond the microbial diversity and community structure assessments (NMDS, ADONIS, ANOSIM), we explored the interactions between the microbial communities coexisting in sponges using the checkerboard score (C-score). Analyses of the microbial association pattern (co-occurrence) among intertidal sympatric sponges revealed the random association of microbes, favoring the hypothesis that the sponge-inhabiting microbes are recruited from the habitat mostly by chance or influenced by environmental factors to benefit the hosts.     

Settlement Patterns of Mediterranean Fruit Flies in the Tree Canopy: An Experimental Analysis

Males of the Mediterranean fruit fly, Ceratitis capitata (medfly), defend individual leaves of host and non-host trees as mating territories and attract females via production of a sex pheromone (a behavior termed ‘calling’). Males tend to aggregate within the tree canopy, and the present study examined the influence of (1) location and (2) resources on female visitation to male aggregations. Males were placed in screen-covered containers, which were suspended on field-caged trees at two locations selected to differ in only one of the following parameters: height above ground (low versus high), leaf density (sparse versus dense), distance to main trunk (canopy edge versus center), proximity to a food source (bird feces absent or present), and proximity to a host fruit (oranges absent or present). Females were released, and we periodically recorded the number of calling males in each of the test aggregations as well as the number of females sighted within 15 cm of each male aggregation. Male calling did not differ between treatments in any of the experiments. However, females were more frequently sighted near males calling (1) in dense foliage, (2) near the central trunk, (3) near bird feces, and (4) near oranges. In an additional experiment, we compared male settlement (in the absence of females) on leaves in the same sites used previously to determine the degree of congruence between the sexes in microhabitat use. Aside from the lack of center-edge effect, males displayed the same settlement patterns as females. In a final experiment, we found no effect of male calling on the distribution of other males: released males were sighted with equal frequency near cups containing mature, calling males and cups containing immature, non-calling males. Medfly lek behavior is discussed in light of these findings.     

Simultaneous Amplicon Sequencing to Explore Co-Occurrence Patterns of Bacterial,Archaeal and Eukaryotic Microorganisms in Rumen Microbial Communities

Ruminants rely on a complex rumen microbial community to convert dietary plant material to energy-yielding products. Here we developed a method to simultaneously analyze the community''s bacterial and archaeal 16S rRNA genes, ciliate 18S rRNA genes and anaerobic fungal internal transcribed spacer 1 genes using 12 DNA samples derived from 11 different rumen samples from three host species (Ovis aries, Bos taurus, Cervus elephas) and multiplex 454 Titanium pyrosequencing. We show that the mixing ratio of the group-specific DNA templates before emulsion PCR is crucial to compensate for differences in amplicon length. This method, in contrast to using a non-specific universal primer pair, avoids sequencing non-targeted DNA, such as plant- or endophyte-derived rRNA genes, and allows increased or decreased levels of community structure resolution for each microbial group as needed. Communities analyzed with different primers always grouped by sample origin rather than by the primers used. However, primer choice had a greater impact on apparent archaeal community structure than on bacterial community structure, and biases for certain methanogen groups were detected. Co-occurrence analysis of microbial taxa from all three domains of life suggested strong within- and between-domain correlations between different groups of microorganisms within the rumen. The approach used to simultaneously characterize bacterial, archaeal and eukaryotic components of a microbiota should be applicable to other communities occupying diverse habitats.     

An Analysis of Citizen Science Based Research: Usage and Publication Patterns

The use of citizen science for scientific discovery relies on the acceptance of this method by the scientific community. Using the Web of Science and Scopus as the source of peer reviewed articles, an analysis of all published articles on “citizen science” confirmed its growth, and found that significant research on methodology and validation techniques preceded the rapid rise of the publications on research outcomes based on citizen science methods. Of considerable interest is the growing number of studies relying on the re-use of collected datasets from past citizen science research projects, which used data from either individual or multiple citizen science projects for new discoveries, such as for climate change research. The extent to which citizen science has been used in scientific discovery demonstrates its importance as a research approach. This broad analysis of peer reviewed papers on citizen science, that included not only citizen science projects, but the theory and methods developed to underpin the research, highlights the breadth and depth of the citizen science approach and encourages cross-fertilization between the different disciplines.     

Universal Patterns in Cultural Evolution: An Empirical Analysis Using Guttman Scaling

We test for universal patterns in cultural evolution by Guttman scaling on two different worldwide samples of archaeological traditions and on well-known archaeological sequences. The evidence is generally consistent with universal evolutionary sequences. We also present evidence for some punctuated evolutionary events.     

Bison Movement Patterns On The Canadian Plains: An Ecological Analysis

Abstract

The primary object of this study is to determine whether bison movements were seasonal and sufficiently regular to be considered migratory.     

Patterns of Information-Seeking for Cancer on the Internet: An Analysis of Real World Data

Although traditionally the primary information sources for cancer patients have been the treating medical team, patients and their relatives increasingly turn to the Internet, though this source may be misleading and confusing. We assess Internet searching patterns to understand the information needs of cancer patients and their acquaintances, as well as to discern their underlying psychological states. We screened 232,681 anonymous users who initiated cancer-specific queries on the Yahoo Web search engine over three months, and selected for study users with high levels of interest in this topic. Searches were partitioned by expected survival for the disease being searched. We compared the search patterns of anonymous users and their contacts. Users seeking information on aggressive malignancies exhibited shorter search periods, focusing on disease- and treatment-related information. Users seeking knowledge regarding more indolent tumors searched for longer periods, alternated between different subjects, and demonstrated a high interest in topics such as support groups. Acquaintances searched for longer periods than the proband user when seeking information on aggressive (compared to indolent) cancers. Information needs can be modeled as transitioning between five discrete states, each with a unique signature representing the type of information of interest to the user. Thus, early phases of information-seeking for cancer follow a specific dynamic pattern. Areas of interest are disease dependent and vary between probands and their contacts. These patterns can be used by physicians and medical Web site authors to tailor information to the needs of patients and family members.     

The Surname Space of the Czech Republic: Examining Population Structure by Network Analysis of Spatial Co-Occurrence of Surnames

In the majority of countries, surnames represent a ubiquitous cultural attribute inherited from an individual''s ancestors and predominantly only altered through marriage. This paper utilises an innovative method, taken from economics, to offer unprecedented insights into the “surname space” of the Czech Republic. We construct this space as a network based on the pairwise probabilities of co-occurrence of surnames and find that the network representation has clear parallels with various ethno-cultural boundaries in the country. Our inductive approach therefore formalizes a simple assumption that the more frequently the bearers of two surnames concentrate in the same locations the higher the probability that these two surnames can be related (considering ethno-cultural relatedness, common co-ancestry or genetic relatedness, or some other type of relatedness). Using the Czech Republic as a case study this paper offers a fresh perspective on surnames as a quantitative data source and provides a methodology that can be easily incorporated within wider cultural, ethnic, geographic and population genetics studies already utilizing surnames.     

Inferred vs Realized Patterns of Gene Flow: An Analysis of Population Structure in the Andros Island Rock Iguana

Ecological data, the primary source of information on patterns and rates of migration, can be integrated with genetic data to more accurately describe the realized connectivity between geographically isolated demes. In this paper we implement this approach and discuss its implications for managing populations of the endangered Andros Island Rock Iguana, Cyclura cychlura cychlura. This iguana is endemic to Andros, a highly fragmented landmass of large islands and smaller cays. Field observations suggest that geographically isolated demes were panmictic due to high, inferred rates of gene flow. We expand on these observations using 16 polymorphic microsatellites to investigate the genetic structure and rates of gene flow from 188 Andros Iguanas collected across 23 island sites. Bayesian clustering of specimens assigned individuals to three distinct genotypic clusters. An analysis of molecular variance (AMOVA) indicates that allele frequency differences are responsible for a significant portion of the genetic variance across the three defined clusters (F_st =  0.117, p0.01). These clusters are associated with larger islands and satellite cays isolated by broad water channels with strong currents. These findings imply that broad water channels present greater obstacles to gene flow than was inferred from field observation alone. Additionally, rates of gene flow were indirectly estimated using BAYESASS 3.0. The proportion of individuals originating from within each identified cluster varied from 94.5 to 98.7%, providing further support for local isolation. Our assessment reveals a major disparity between inferred and realized gene flow. We discuss our results in a conservation perspective for species inhabiting highly fragmented landscapes.     

Neighborhood Socioeconomic Circumstances and the Co-Occurrence of Unhealthy Lifestyles: Evidence from 206,457 Australians in the 45 and Up Study

Background

Research on the co-occurrence of unhealthy lifestyles has tended to focus mainly upon the demographic and socioeconomic characteristics of individuals. This study investigated the relevance of neighborhood socioeconomic circumstance for multiple unhealthy lifestyles.

Method

An unhealthy lifestyle index was constructed for 206,457 participants in the 45 and Up Study (2006–2009) by summing binary responses on smoking, alcohol, physical activity and five diet-related variables. Higher scores indicated the co-occurrence of unhealthy lifestyles. Association with self-rated health, quality of life; and risk of psychological distress was investigated using multilevel logistic regression. Association between the unhealthy lifestyle index with neighborhood characteristics (local affluence and geographic remoteness) were assessed using multilevel linear regression, adjusting for individual-level characteristics.

Results

Nearly 50% of the sample reported 3 or 4 unhealthy lifestyles. Only 1.5% reported zero unhealthy lifestyles and 0.2% had all eight. Compared to people who scored zero, those who scored 8 (the ‘unhealthiest’ group) were 7 times more likely to rate their health as poor (95%CI 3.6, 13.7), 5 times more likely to report poor quality of life (95%CI 2.6, 10.1), and had a 2.6 times greater risk of psychological distress (95%CI 1.8, 3.7). Higher scores among men decreased with age, whereas a parabolic distribution was observed among women. Neighborhood affluence was independently associated with lower scores on the unhealthy lifestyle index. People on high incomes scored higher on the unhealthy lifestyle index if they were in poorer neighborhoods, while those on low incomes had fewer unhealthy lifestyles if living in more affluent areas.

Interpretation

Residents of deprived neighborhoods tend to report more unhealthy lifestyles than their peers in affluent areas, regardless of their individual demographic and socioeconomic characteristics. Future research should investigate the trade-offs of population-level versus geographically targeted multiple lifestyle interventions.     

Co-Occurrence of Preference Functions and Acceptance Thresholds in Female Choice: Mate Discrimination in the Lesser Wax Moth

Basic economic models adapted from foraging theory predict that decisions in mate choice may be determined either by ‘best‐of‐n’ preference functions or by sequential rules incorporating acceptance thresholds. However, in some species, more complex determinations that incorporate versions of both protocols are found. To understand the functions of co‐occurring protocols, we studied mating decisions in the lesser wax moth, Achroia grisella (Lepidoptera: Pyralidae), an acoustic species in which females prefer males, the advertisement songs of which are delivered at relatively high ‘pulse‐pair’ rates. In addition to this preference, A. grisella females avoid mating with a male, the song of which does not exceed a minimum pulse‐pair rate, and they hold to this criterion even when no other singing males are present and regardless of song amplitude. Thus, mating decisions are not simply based on acoustic power (pulse‐pair rate × amplitude). We recorded male songs and female responses in an A. grisella population and found that male pulse‐pair rates showed a median of 87/s and ranged from 50 to 115/s, while female acceptance thresholds for male song showed a median of 60/s and ranged from 30 to 105/s. The distributions of thresholds were approximately normal and were not significantly skewed toward the right. Male song rates declined slightly with age, but female thresholds remained stable over the adult lifespan. Both the male and female traits showed significant repeatability within individuals. Whereas phylogenetic inference indicates that hearing in pyralid moths originated as a means of avoiding predation by insectivorous bats, the specific distribution of female acceptance thresholds suggests that currently this protocol does not primarily function to preclude inappropriate, and potentially lethal, responses to bat echolocations: pulse rates in the searching‐phase echolocations used by either aerial‐hawking or substrate‐gleaning bats mostly range from 10 to 20/s, and the lack of positive skew in the distribution of thresholds indicates an absence of directional selection from the left. Rather, we infer that thresholds augment preference functions in A. grisella by precluding mating with males which are markedly inferior in a critical song character. In general, co‐occurring protocols may be important where population density fluctuates markedly, as preference functions may be ineffective in preventing mating with inferior males when density is low.     

Patterns of Human Papillomavirus Types in Multiple Infections: An Analysis in Women and Men of the High Throughput Human Papillomavirus Monitoring Study

Background

To evaluate the pattern of co-infection of human papillomavirus (HPV) types in both sexes in Sweden.

Methods

Cell samples from genital swabs, first-void urine, and genital swabs immersed in first-void urine were collected in the present cross-sectional High Throughput HPV Monitoring study. Overall, 31,717 samples from women and 9,949 from men (mean age 25) were tested for 16 HPV types using mass spectrometry. Multilevel logistic regression was used to estimate the expected number of multiple infections with specific HPV types, adjusted for age, type of sample, and accounting for correlations between HPV types due to unobserved risk factors using sample-level random effects. Bonferroni correction was used to allow for multiple comparisons (120).

Results

Observed-to-expected ratio for any multiple infections was slightly above unity in both sexes, but, for most 2-type combinations, there was no evidence of significant departure from expected numbers. HPV6/18 was found more often and HPV51/68 and 6/68 less often than expected. However, HPV68 tended to be generally underrepresented in co-infections, suggesting a sub-optimal performance of our testing method for this HPV type.

Conclusions

We found no evidence for positive or negative clustering between HPV types included in the current prophylactic vaccines and other untargeted oncogenic types, in either sex.