Species Identity, Life History, and Geographic Distance Influence Gut Bacterial Communities in Lab-Reared and European Field-Collected Culicoides Biting midges.

Bacteria are part of the insect gut system and influence many physiological traits of their host. Gut bacteria may even reduce or block the transmission of arboviruses in several species of arthropod vectors. Culicoides biting midges are important arboviral vectors of several livestock and wildlife diseases, yet limited information is available on their gut bacterial communities. Addressing this gap will help inform how these communities can be manipulated and ultimately used as novel tools to control pathogens. To assess how bacterial communities change during the life stages of lab-reared C. nubeculosus and C. sonorensis, endosymbiotic bacteria were identified using Illumina sequencing of 16S rRNA and taxonomically characterised. Analyses were conducted to determine how gut bacterial communities in adults are influenced by species identity and geographic distance among biting midge populations. Communities of the two lab-reared Culicoides species significantly changed after pupation and with maturation into 6-day-old adults. Pseudomonas, Burkholderiaceae and Leucobacter bacteria were part of a core community that was trans-stadially transmitted and found throughout their life cycle. Among field-collected biting midges, the bacterial communities were unique for almost each species. Cardinium, Rickettsia and Wolbachia were some of the most abundant bacteria in midges collected from wetlands. Only Pseudomonas was present in high relative abundance in all field-collected species. In this study, species identity, as well as geographic distance, influenced the gut bacterial communities and may partly explain known inter- and intra-species variability in vector competence. Additionally, stably associated bacterial species could be candidates for paratransgenic strategies to control vector-borne pathogens.

Log-ratio analysis of microbiome data with many zeroes is library size dependent.

Microbiome composition data collected through amplicon sequencing are count data on taxa in which the total count per sample (the library size) is an artefact of the sequencing platform, and as a result, such data are compositional. To avoid library size dependency, one common way of analysing multivariate compositional data is to perform a principal component analysis (PCA) on data transformed with the centred log-ratio, hereafter called a log-ratio PCA. Two aspects typical of amplicon sequencing data are the large differences in library size and the large number of zeroes. In this study, we show on real data and by simulation that, applied to data that combine these two aspects, log-ratio PCA is nevertheless heavily dependent on the library size. This leads to a reduction in power when testing against any explanatory variable in log-ratio redundancy analysis. If there is additionally a correlation between the library size and the explanatory variable, then the type 1 error becomes inflated. We explore putative solutions to this problem.

Double constrained ordination for assessing biological trait responses to multiple stressors: A case study with benthic macroinvertebrate communities.

Benthic macroinvertebrate communities are used as indicators for anthropogenic stress in freshwater ecosystems. To better understand the relationship between anthropogenic stress and changes in macroinvertebrate community composition, it is important to understand how different stressors and species traits are associated, and how these associations influence variation in species occurrence and abundances. Here, we show the capacity of the multivariate technique of double constrained correspondence analysis (dc-CA) to analyse trait-environment relationships, and we compare it with the redundancy analysis method on community weighted mean values of traits (CWM-RDA), which is frequently used for this type of analysis. The analyses were based on available biomonitoring data for macroinvertebrate communities from the Danube River. Results from forward selection of traits and environmental variables using dc-CA analyses showed that aquatic stages, reproduction techniques, dispersal tactics, locomotion and substrate relations, altitude, longitudinal and transversal distribution, and substrate preferendum were significantly related to habitat characteristics, hydromorphological alterations and water quality measurements such as physico-chemical parameters, heavy metals, pesticides and pharmaceuticals. Environmental variables significantly associated with traits using the CWM-RDA method were generally consistent with those found in dc-CA analysis. However, the CWM-RDA does neither test nor explicitly select traits, while dc-CA tests and selects both traits and environmental variables. Moreover, the dc-CA analysis revealed that the set of environmental variables was much better in explaining the community data than the available trait set, a kind of information that can neither be obtained from CWM-RDA nor from RLQ (Environment, Link and Trait data), which is a close cousin of dc-CA but not regression-based. Our results suggest that trait-based analysis based on dc-CA may be useful to assess mechanistic links between multiple anthropogenic stressors and ecosystem health, but more data sets should be analysed in the same manner.

An improved statistical approach for reconstructing past climates from biotic assemblages.

Quantitative reconstructions of past climates are an important resource for evaluating how well climate models reproduce climate changes. One widely used statistical approach for making such reconstructions from fossil biotic assemblages is weighted averaging partial least-squares regression (WA-PLS). There is however a known tendency for WA-PLS to yield reconstructions compressed towards the centre of the climate range used for calibration, potentially biasing the reconstructed past climates. We present an improvement of WA-PLS by assuming that: (i) the theoretical abundance of each taxon is unimodal with respect to the climate variable considered; (ii) observed taxon abundances follow a multinomial distribution in which the total abundance of a sample is climatically uninformative; and (iii) the estimate of the climate value at a given site and time makes the observation most probable, i.e. it maximizes the log-likelihood function. This climate estimate is approximated by weighting taxon abundances in WA-PLS by the inverse square of their climate tolerances. We further improve the approach by considering the frequency ( fx) of the climate variable in the training dataset. Tolerance-weighted WA-PLS with fx correction greatly reduces the compression bias, compared with WA-PLS, and improves model performance in reconstructions based on an extensive modern pollen dataset.

Impact of Gut Bacteria on the Infection and Transmission of Pathogenic Arboviruses by Biting Midges and Mosquitoes.

Tripartite interactions among insect vectors, midgut bacteria, and viruses may determine the ability of insects to transmit pathogenic arboviruses. Here, we investigated the impact of gut bacteria on the susceptibility of Culicoides nubeculosus and Culicoides sonorensis biting midges for Schmallenberg virus, and of Aedes aegypti mosquitoes for Zika and chikungunya viruses. Gut bacteria were manipulated by treating the adult insects with antibiotics. The gut bacterial communities were investigated using Illumina MiSeq sequencing of 16S rRNA, and susceptibility to arbovirus infection was tested by feeding insects with an infectious blood meal. Antibiotic treatment led to changes in gut bacteria for all insects. Interestingly, the gut bacterial composition of untreated Ae. aegypti and C. nubeculosus showed Asaia as the dominant genus, which was drastically reduced after antibiotic treatment. Furthermore, antibiotic treatment resulted in relatively more Delftia bacteria in both biting midge species, but not in mosquitoes. Antibiotic treatment and subsequent changes in gut bacterial communities were associated with a significant, 1.8-fold increased infection rate of C. nubeculosus with Schmallenberg virus, but not for C. sonorensis. We did not find any changes in infection rates for Ae. aegypti mosquitoes with Zika or chikungunya virus. We conclude that resident gut bacteria may dampen arbovirus transmission in biting midges, but not so in mosquitoes. Use of antimicrobial compounds at livestock farms might therefore have an unexpected contradictory effect on the health of animals, by increasing the transmission of viral pathogens by biting midges.

Benthic invertebrate and microbial biodiversity in sub-tropical urban rivers: Correlations with environmental variables and emerging chemicals.

Urban rivers often function as sinks for various contaminants potentially placing the benthic communities at risk of exposure. We performed a comprehensive biological survey of the benthic macroinvertebrate and bacterial community compositions in six rivers from the suburb to the central urban area of Guangzhou city (South China), and evaluated their correlations with emerging organic contaminants, heavy metals and nutrients. Overall, the benthic macroinvertebrate community shifted from molluscs to oligochaete from the suburban to the central urban rivers that receive treated and untreated sewage. An exception was the site in the Sha River where chironomids were most abundant. The differences in macroinvertebrate community assemblages were significantly associated with chromium, total phosphorus, galaxolide, triclosan and sand content in the sediment. There was no significant difference in benthic macroinvertebrate composition between the dry and wet season. As assessed by double constrained ordination, sexual reproduction was the only trait of benthic macroinvertebrates that showed a significant correlation with pollution variables, as it was significantly positively correlated with chromium and total phosphorus. This suggests that r-strategist occurs in polluted sampling sites. The benthic bacterial community composition showed a significant difference between seasons and among the Liuxi River, Zhujiang River and central urban rivers. The differences in community composition of the benthic bacteria were significantly correlated with galaxolide, total phosphorus, lead and triclosan. These results suggest that input of treated and untreated sewage significantly altered the benthic macroinvertebrate and bacterial community compositions in urban rivers.

Investigating microbial associations from sequencing survey data with co-correspondence analysis.

Microbial communities, which drive major ecosystem functions, consist of a wide range of interacting species. Understanding how microbial communities are structured and the processes underlying this is crucial to interpreting ecosystem responses to global change but is challenging as microbial interactions cannot usually be directly observed. Multiple efforts are currently focused to combine next-generation sequencing (NGS) techniques with refined statistical analysis (e.g., network analysis, multivariate analysis) to characterize the structures of microbial communities. However, most of these approaches consider a single table of sequencing data measured for several samples. Technological advances now make it possible to collect NGS data on different taxonomic groups simultaneously for the same samples, allowing us to analyse a pair of tables. Here, an analytical framework based on co-correspondence analysis (CoCA) is proposed to study the distributions, assemblages and interactions between two microbial communities. We show the ability of this approach to highlight the relationships between two microbial communities, using two data sets exhibiting various types of interactions. CoCA identified strong association patterns between autotrophic and heterotrophic microbial eukaryote assemblages, on the one hand, and between microalgae and viruses, on the other. We demonstrate also how CoCA can be used, complementary to network analysis, to reorder co-occurrence networks and thus investigate the presence of patterns in ecological networks.

Differently Pre-treated Alfalfa Silages Affect the in vitro Ruminal Microbiota Composition.

Alfalfa (Medicago sativa L.) silage (AS) is an important feedstuff in ruminant nutrition. However, its high non-protein nitrogen content often leads to poor ruminal nitrogen retention. Various pre-ensiling treatments differing with respect to dry matter concentrations, wilting intensities and sucrose addition have been previously shown to improve the quality and true protein preservation of AS, and have substantial effects on in vitro ruminal fermentation of the resulting silages. However, it is unknown how these pre-ensiling treatments affect the ruminal microbiota composition, and whether alterations in the microbiota explain previously observed differences in ruminal fermentation. Therefore, during AS incubation in a rumen simulation system, liquid and solid phases were sampled 2 and 7 days after first incubating AS, representing an early (ET) and late (LT) time point, respectively. Subsequently, DNA was extracted and qPCR (bacteria, archaea, and anaerobic fungi) and prokaryotic 16S rRNA gene amplicon sequence analyses were performed. At the ET, high dry matter concentration and sucrose addition increased concentrations of archaea in the liquid phase (P = 0.001) and anaerobic fungi in the solid phase (P < 0.001). At the LT, only sucrose addition increased archaeal concentration in the liquid phase (P = 0.014) and anaerobic fungal concentration in the solid phase (P < 0.001). Bacterial concentrations were not affected by pre-ensiling treatments. The prokaryotic phylogenetic diversity index decreased in the liquid phase from ET to LT (P = 0.034), whereas the solid phase was not affected (P = 0.060). This is suggestive of a general adaption of the microbiota to the soluble metabolites released from the incubated AS, particularly regarding the sucrose-treated AS. Redundancy analysis of the sequence data at the genus level indicated that sucrose addition (P = 0.001), time point (P = 0.001), and their interaction (P = 0.001) affected microbial community composition in both phases. In summary, of the pre-ensiling treatments tested sucrose addition had the largest effect on the microbiota, and together with sampling time point affected microbiota composition in both phases of the rumen simulation system. Thus, microbiota composition analysis helped to understand the ruminal fermentation patterns, but could not fully explain them.

Biomarker Research in ADHD: the Impact of Nutrition (BRAIN) - study protocol of an open-label trial to investigate the mechanisms underlying the effects of a few-foods diet on ADHD symptoms in children.

INTRODUCTION: Attention deficit hyperactivity disorder (ADHD) is the most common childhood behavioural disorder, causing significant impediment to a child's development. It is a complex disorder with numerous contributing (epi)genetic and environmental factors. Currently, treatment consists of behavioural and pharmacological therapy. However, ADHD medication is associated with several side effects, and concerns about long-term effects and efficacy exist. Therefore, there is considerable interest in the development of alternative treatment options. Double-blind research investigating the effects of a few-foods diet (FFD) has demonstrated a significant decrease in ADHD symptoms following an FFD. However, an FFD requires a considerable effort of both child and parents, limiting its applicability as a general ADHD treatment. To make FFD intervention less challenging or potentially obsolete, we need to understand how, and in which children, an FFD affects ADHD behaviour and, consequently, the child's well-being. We hypothesise that an FFD affects brain function, and that the nutritional impact on ADHD is effectuated by a complex interplay between the microbiota, gut and brain, that is, the microbiota-gut-brain axis. METHODS AND ANALYSIS: The Biomarker Research in ADHD: the Impact of Nutrition (BRAIN) study is an open-label trial with researchers blinded to changes in ADHD symptoms during sample processing and initial data analyses. ETHICS AND DISSEMINATION: The Medical Research and Ethics Committee of Wageningen University has approved this study (NL63851.081.17, application 17/24). Results will be disseminated through peer-reviewed journal publications, conference presentations, (social) media and the BRAIN study website. A summary of the findings will be provided to the participants. TRIAL REGISTRATION NUMBER: NCT03440346. STUDY DATES: Collection of primary outcome data started in March 2018 and will be ongoing until 100 children have participated in the study. Sample data analysis will start after all samples have been collected.

Integrating spatial and phylogenetic information in the fourth-corner analysis to test trait-environment relationships.

The fourth-corner analysis aims to quantify and test for relationships between species traits and site-specific environmental variables, mediated by site-specific species abundances. Since there is no common unit of observation, the significance of the relationships is tested using a double permutation procedure (site based and species based). This method implies that all species and sites are independent of each other. However, this fundamental hypothesis might be flawed because of phylogenetic relatedness between species and spatial autocorrelation in the environmental data. Here, using a simulation-based experiment, we demonstrate how the presence of spatial and phylogenetic autocorrelations can, in some circumstances, lead to inflated type I error rates, suggesting that significant associations can be misidentified. As an alternative, we propose a new randomization approach designed to avoid this issue, based on Moran's spectral randomization. In this approach, standard permutations are replaced by constrained randomizations so that the distribution of the statistic under the null hypothesis is built with additional constraints to preserve the phylogenetic and spatial structures of the observed data. The inclusion of this new randomization approach provides total control over type I error rates and should be used in real studies where spatial and phylogenetic autocorrelations often occur.

A critical issue in model-based inference for studying trait-based community assembly and a solution.

Statistical testing of trait-environment association from data is a challenge as there is no common unit of observation: the trait is observed on species, the environment on sites and the mediating abundance on species-site combinations. A number of correlation-based methods, such as the community weighted trait means method (CWM), the fourth-corner correlation method and the multivariate method RLQ, have been proposed to estimate such trait-environment associations. In these methods, valid statistical testing proceeds by performing two separate resampling tests, one site-based and the other species-based and by assessing significance by the largest of the two p-values (the pmax test). Recently, regression-based methods using generalized linear models (GLM) have been proposed as a promising alternative with statistical inference via site-based resampling. We investigated the performance of this new approach along with approaches that mimicked the pmax test using GLM instead of fourth-corner. By simulation using models with additional random variation in the species response to the environment, the site-based resampling tests using GLM are shown to have severely inflated type I error, of up to 90%, when the nominal level is set as 5%. In addition, predictive modelling of such data using site-based cross-validation very often identified trait-environment interactions that had no predictive value. The problem that we identify is not an "omitted variable bias" problem as it occurs even when the additional random variation is independent of the observed trait and environment data. Instead, it is a problem of ignoring a random effect. In the same simulations, the GLM-based pmax test controlled the type I error in all models proposed so far in this context, but still gave slightly inflated error in more complex models that included both missing (but important) traits and missing (but important) environmental variables. For screening the importance of single trait-environment combinations, the fourth-corner test is shown to give almost the same results as the GLM-based tests in far less computing time.

Combining exposure and effect modeling into an integrated probabilistic environmental risk assessment for nanoparticles.

There is a growing need for good environmental risk assessment of engineered nanoparticles (ENPs). Environmental risk assessment of ENPs has been hampered by lack of data and knowledge about ENPs, their environmental fate, and their toxicity. This leads to uncertainty in the risk assessment. To deal with uncertainty in the risk assessment effectively, probabilistic methods are advantageous. In the present study, the authors developed a method to model both the variability and the uncertainty in environmental risk assessment of ENPs. This method is based on the concentration ratio and the ratio of the exposure concentration to the critical effect concentration, both considered to be random. In this method, variability and uncertainty are modeled separately so as to allow the user to see which part of the total variation in the concentration ratio is attributable to uncertainty and which part is attributable to variability. The authors illustrate the use of the method with a simplified aquatic risk assessment of nano-titanium dioxide. The authors' method allows a more transparent risk assessment and can also direct further environmental and toxicological research to the areas in which it is most needed. Environ Toxicol Chem 2016;35:2958-2967. © 2016 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals, Inc. on behalf of SETAC.

Parametric estimation of P(X > Y) for normal distributions in the context of probabilistic environmental risk assessment.

Estimating the risk, P(X > Y), in probabilistic environmental risk assessment of nanoparticles is a problem when confronted by potentially small risks and small sample sizes of the exposure concentration X and/or the effect concentration Y. This is illustrated in the motivating case study of aquatic risk assessment of nano-Ag. A non-parametric estimator based on data alone is not sufficient as it is limited by sample size. In this paper, we investigate the maximum gain possible when making strong parametric assumptions as opposed to making no parametric assumptions at all. We compare maximum likelihood and Bayesian estimators with the non-parametric estimator and study the influence of sample size and risk on the (interval) estimators via simulation. We found that the parametric estimators enable us to estimate and bound the risk for smaller sample sizes and small risks. Also, the Bayesian estimator outperforms the maximum likelihood estimators in terms of coverage and interval lengths and is, therefore, preferred in our motivating case study.

Integrated probabilistic risk assessment for nanoparticles: the case of nanosilica in food.

Insight into risks of nanotechnology and the use of nanoparticles is an essential condition for the social acceptance and safe use of nanotechnology. One of the problems with which the risk assessment of nanoparticles is faced is the lack of data, resulting in uncertainty in the risk assessment. We attempt to quantify some of this uncertainty by expanding a previous deterministic study on nanosilica (5-200 nm) in food into a fully integrated probabilistic risk assessment. We use the integrated probabilistic risk assessment method in which statistical distributions and bootstrap methods are used to quantify uncertainty and variability in the risk assessment. Due to the large amount of uncertainty present, this probabilistic method, which separates variability from uncertainty, contributed to a better understandable risk assessment. We found that quantifying the uncertainties did not increase the perceived risk relative to the outcome of the deterministic study. We pinpointed particular aspects of the hazard characterization that contributed most to the total uncertainty in the risk assessment, suggesting that further research would benefit most from obtaining more reliable data on those aspects.

Analysing chemical-induced changes in macroinvertebrate communities in aquatic mesocosm experiments: a comparison of methods.

Mesocosm experiments that study the ecological impact of chemicals are often analysed using the multivariate method 'Principal Response Curves' (PRCs). Recently, the extension of generalised linear models (GLMs) to multivariate data was introduced as a tool to analyse community data in ecology. Moreover, data aggregation techniques that can be analysed with univariate statistics have been proposed. The aim of this study was to compare their performance. We compiled macroinvertebrate abundance datasets of mesocosm experiments designed for studying the effect of various organic chemicals, mainly pesticides, and re-analysed them. GLMs for multivariate data and selected aggregated endpoints were compared to PRCs regarding their performance and potential to identify affected taxa. In addition, we analysed the inter-replicate variability encountered in the studies. Mesocosm experiments characterised by a higher taxa richness of the community and/or lower taxonomic resolution showed a greater inter-replicate variability, whereas variability decreased the more zero counts were encountered in the samples. GLMs for multivariate data performed equally well as PRCs regarding the community response. However, compared to first axis PRCs, GLMs provided a better indication of individual taxa responding to treatments, as separate models are fitted to each taxon. Data aggregation methods performed considerably poorer compared to PRCs. Multivariate community data, which are generated during mesocosm experiments, should be analysed using multivariate methods to reveal treatment-related community-level responses. GLMs for multivariate data are an alternative to the widely used PRCs.

A unimodal species response model relating traits to environment with application to phytoplankton communities.

In this paper we attempt to explain observed niche differences among species (i.e. differences in their distribution along environmental gradients) by differences in trait values (e.g. volume) in phytoplankton communities. For this, we propose the trait-modulated Gaussian logistic model in which the niche parameters (optimum, tolerance and maximum) are made linearly dependent on species traits. The model is fitted to data in the Bayesian framework using OpenBUGS (Bayesian inference Using Gibbs Sampling) to identify according to which environmental variables there is niche differentiation among species and traits. We illustrate the method with phytoplankton community data of 203 lakes located within four climate zones and associated measurements on 11 environmental variables and six morphological species traits of 60 species. Temperature and chlorophyll-a (with opposite signs) described well the niche structure of all species. Results showed that about 25% of the variance in the niche centres with respect to chlorophyll-a were accounted for by traits, whereas niche width and maximum could not be predicted by traits. Volume, mucilage, flagella and siliceous exoskeleton are found to be the most important traits to explain the niche centres. Species were clustered in two groups with different niches structures, group 1 high temperature-low chlorophyll-a species and group 2 low temperature-high chlorophyll-a species. Compared to group 2, species in group 1 had larger volume but lower surface area, had more often flagella but neither mucilage nor siliceous exoskeleton. These results might help in understanding the effect of environmental changes on phytoplankton community. The proposed method, therefore, can also apply to other aquatic or terrestrial communities for which individual traits and environmental conditioning factors are available.

Combining the fourth-corner and the RLQ methods for assessing trait responses to environmental variation.

Assessing trait responses to environmental gradients requires the simultaneous analysis of the information contained in three tables: L (species distribution across samples), R (environmental characteristics of samples), and Q (species traits). Among the available methods, the so-called fourth-corner and RLQ methods are two appealing alternatives that provide a direct way to test and estimate trait-nvironment relationships. Both methods are based on the analysis of the fourth-corner matrix, which crosses traits and environmental variables weighted by species abundances. However, they differ greatly in their outputs: RLQ is a multivariate technique that provides ordination scores to summarize the joint structure among the three tables, whereas the fourth-corner method mainly tests for individual trait-environment relationships (i.e., one trait and one environmental variable at a time). Here, we illustrate how the complementarity between these two methods can be exploited to promote new ecological knowledge and to improve the study of trait-environment relationships. After a short description of each method, we apply them to real ecological data to present their different outputs and provide hints about the gain resulting from their combined use.

Reset of a critically disturbed microbial ecosystem: faecal transplant in recurrent Clostridium difficile infection.

Recurrent Clostridium difficile infection (CDI) can be effectively treated by infusion of a healthy donor faeces suspension. However, it is unclear what factors determine treatment efficacy. By using a phylogenetic microarray platform, we assessed composition, diversity and dynamics of faecal microbiota before, after and during follow-up of the transplantation from a healthy donor to different patients, to elucidate the mechanism of action of faecal infusion. Global composition and network analysis of the microbiota was performed in faecal samples from nine patients with recurrent CDI. Analyses were performed before and after duodenal donor faeces infusion, and during a follow-up of 10 weeks. The microbiota data were compared with that of the healthy donors. All patients successfully recovered. Their intestinal microbiota changed from a low-diversity diseased state, dominated by Proteobacteria and Bacilli, to a more diverse ecosystem resembling that of healthy donors, dominated by Bacteroidetes and Clostridium groups, including butyrate-producing bacteria. We identified specific multi-species networks and signature microbial groups that were either depleted or restored as a result of the treatment. The changes persisted over time. Comprehensive and deep analyses of the microbiota of patients before and after treatment exposed a therapeutic reset from a diseased state towards a healthy profile. The identification of microbial groups that constitute a niche for C. difficile overgrowth, as well as those driving the reinstallation of a healthy intestinal microbiota, could contribute to the development of biomarkers predicting recurrence and treatment outcome, identifying an optimal microbiota composition that could lead to targeted treatment strategies.

Generalized linear mixed models can detect unimodal species-environment relationships.

Niche theory predicts that species occurrence and abundance show non-linear, unimodal relationships with respect to environmental gradients. Unimodal models, such as the Gaussian (logistic) model, are however more difficult to fit to data than linear ones, particularly in a multi-species context in ordination, with trait modulated response and when species phylogeny and species traits must be taken into account. Adding squared terms to a linear model is a possibility but gives uninterpretable parameters. This paper explains why and when generalized linear mixed models, even without squared terms, can effectively analyse unimodal data and also presents a graphical tool and statistical test to test for unimodal response while fitting just the generalized linear mixed model. The R-code for this is supplied in Supplemental Information 1.

An automated system for the recognition of various specific rat behaviours.

The automated measurement of rodent behaviour is crucial to advance research in neuroscience and pharmacology. Rats and mice are used as models for human diseases; their behaviour is studied to discover and develop new drugs for psychiatric and neurological disorders and to establish the effect of genetic variation on behavioural changes. Such behaviour is primarily labelled by humans. Manual annotation is labour intensive, error-prone and subject to individual interpretation. We present a system for automated behaviour recognition (ABR) that recognises the rat behaviours 'drink', 'eat', 'sniff', 'groom', 'jump', 'rear unsupported', 'rear wall', 'rest', 'twitch' and 'walk'. The ABR system needs no on-site training; the only inputs needed are the sizes of the cage and the animal. This is a major advantage over other systems that need to be trained with hand-labelled data before they can be used in a new experimental setup. Furthermore, ABR uses an overhead camera view, which is more practical in lab situations and facilitates high-throughput testing more easily than a side-view setup. ABR has been validated by comparison with manual behavioural scoring by an expert. For this, animals were treated with two types of psychopharmaca: a stimulant drug (Amphetamine) and a sedative drug (Diazepam). The effects of drug treatment on certain behavioural categories were measured and compared for both analysis methods. Statistical analysis showed that ABR found similar behavioural effects as the human observer. We conclude that our ABR system represents a significant step forward in the automated observation of rodent behaviour.