Author Correction: Reproducibility of animal research in light of biological variation.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Reproducibility of animal research in light of biological variation.

Context-dependent biological variation presents a unique challenge to the reproducibility of results in experimental animal research, because organisms' responses to experimental treatments can vary with both genotype and environmental conditions. In March 2019, experts in animal biology, experimental design and statistics convened in Blonay, Switzerland, to discuss strategies addressing this challenge. In contrast to the current gold standard of rigorous standardization in experimental animal research, we recommend the use of systematic heterogenization of study samples and conditions by actively incorporating biological variation into study design through diversifying study samples and conditions. Here we provide the scientific rationale for this approach in the hope that researchers, regulators, funders and editors can embrace this paradigm shift. We also present a road map towards better practices in view of improving the reproducibility of animal research.

The rearing environment persistently modulates mouse phenotypes from the molecular to the behavioural level.

The phenotype of an organism results from its genotype and the influence of the environment throughout development. Even when using animals of the same genotype, independent studies may test animals of different phenotypes, resulting in poor replicability due to genotype-by-environment interactions. Thus, genetically defined strains of mice may respond differently to experimental treatments depending on their rearing environment. However, the extent of such phenotypic plasticity and its implications for the replicability of research findings have remained unknown. Here, we examined the extent to which common environmental differences between animal facilities modulate the phenotype of genetically homogeneous (inbred) mice. We conducted a comprehensive multicentre study, whereby inbred C57BL/6J mice from a single breeding cohort were allocated to and reared in 5 different animal facilities throughout early life and adolescence, before being transported to a single test laboratory. We found persistent effects of the rearing facility on the composition and heterogeneity of the gut microbial community. These effects were paralleled by persistent differences in body weight and in the behavioural phenotype of the mice. Furthermore, we show that environmental variation among animal facilities is strong enough to influence epigenetic patterns in neurons at the level of chromatin organisation. We detected changes in chromatin organisation in the regulatory regions of genes involved in nucleosome assembly, neuronal differentiation, synaptic plasticity, and regulation of behaviour. Our findings demonstrate that common environmental differences between animal facilities may produce facility-specific phenotypes, from the molecular to the behavioural level. Furthermore, they highlight an important limitation of inferences from single-laboratory studies and thus argue that study designs should take environmental background into account to increase the robustness and replicability of findings.

Systematic assessment of the replicability and generalizability of preclinical findings: Impact of protocol harmonization across laboratory sites.

The influence of protocol standardization between laboratories on their replicability of preclinical results has not been addressed in a systematic way. While standardization is considered good research practice as a means to control for undesired external noise (i.e., highly variable results), some reports suggest that standardized protocols may lead to idiosyncratic results, thus undermining replicability. Through the EQIPD consortium, a multi-lab collaboration between academic and industry partners, we aimed to elucidate parameters that impact the replicability of preclinical animal studies. To this end, 3 experimental protocols were implemented across 7 laboratories. The replicability of results was determined using the distance travelled in an open field after administration of pharmacological compounds known to modulate locomotor activity (MK-801, diazepam, and clozapine) in C57BL/6 mice as a worked example. The goal was to determine whether harmonization of study protocols across laboratories improves the replicability of the results and whether replicability can be further improved by systematic variation (heterogenization) of 2 environmental factors (time of testing and light intensity during testing) within laboratories. Protocols were tested in 3 consecutive stages and differed in the extent of harmonization across laboratories and standardization within laboratories: stage 1, minimally aligned across sites (local protocol); stage 2, fully aligned across sites (harmonized protocol) with and without systematic variation (standardized and heterogenized cohort); and stage 3, fully aligned across sites (standardized protocol) with a different compound. All protocols resulted in consistent treatment effects across laboratories, which were also replicated within laboratories across the different stages. Harmonization of protocols across laboratories reduced between-lab variability substantially compared to each lab using their local protocol. In contrast, the environmental factors chosen to introduce systematic variation within laboratories did not affect the behavioral outcome. Therefore, heterogenization did not reduce between-lab variability further compared to the harmonization of the standardized protocol. Altogether, these findings demonstrate that subtle variations between lab-specific study protocols may introduce variation across independent replicate studies even after protocol harmonization and that systematic heterogenization of environmental factors may not be sufficient to account for such between-lab variation. Differences in replicability of results within and between laboratories highlight the ubiquity of study-specific variation due to between-lab variability, the importance of transparent and fine-grained reporting of methodologies and research protocols, and the importance of independent study replication.

Small energy benefits of in-wake flying in long-duration migratory flights.

During long-distance migrations, some bird species make use of in-wake flying, which should allow them to profit from the upwash produced by another bird. While indirect evidence supports energy saving as the primary benefit of in-wake flying, measurements are still missing. We equipped migrating northern bald ibises (Geronticus eremita) with high-precision global navigation satellite system data loggers to track their position in the flock. We estimated birds' energy expenditure through different proxies, namely dynamic body acceleration (DBA), heart rate and effective wingbeat frequency. During active flapping flight, DBA estimates dropped off when in-wake compared with when not-in-wake. In addition, effective wingbeat frequency decreased, suggesting an increased use of intermittent gliding flight during in-wake periods. Heart rate varied greatly among individuals, with a clear decrease during gliding-corroborating its energy-saving function. Furthermore, we found consistent proof for decreased heart rate during in-wake flying, by up to 4.2%. Hence, we have shown that flying in the wake of another individual reduces birds' DBA, heart rate and effective wingbeat frequency, which could reflect reduced energy requirement.

Evaluation of an Active LF Tracking System and Data Processing Methods for Livestock Precision Farming in the Poultry Sector.

Tracking technologies offer a way to monitor movement of many individuals over long time periods with minimal disturbances and could become a helpful tool for a variety of uses in animal agriculture, including health monitoring or selection of breeding traits that benefit welfare within intensive cage-free poultry farming. Herein, we present an active, low-frequency tracking system that distinguishes between five predefined zones within a commercial aviary. We aimed to evaluate both the processed and unprocessed datasets against a "ground truth" based on video observations. The two data processing methods aimed to filter false registrations, one with a simple deterministic approach and one with a tree-based classifier. We found the unprocessed data accurately determined birds' presence/absence in each zone with an accuracy of 99% but overestimated the number of transitions taken by birds per zone, explaining only 23% of the actual variation. However, the two processed datasets were found to be suitable to monitor the number of transitions per individual, accounting for 91% and 99% of the actual variation, respectively. To further evaluate the tracking system, we estimated the error rate of registrations (by applying the classifier) in relation to three factors, which suggested a higher number of false registrations towards specific areas, periods with reduced humidity, and periods with reduced temperature. We concluded that the presented tracking system is well suited for commercial aviaries to measure individuals' transitions and individuals' presence/absence in predefined zones. Nonetheless, under these settings, data processing remains a necessary step in obtaining reliable data. For future work, we recommend the use of automatic calibration to improve the system's performance and to envision finer movements.

Reproducibility of preclinical animal research improves with heterogeneity of study samples.

Single-laboratory studies conducted under highly standardized conditions are the gold standard in preclinical animal research. Using simulations based on 440 preclinical studies across 13 different interventions in animal models of stroke, myocardial infarction, and breast cancer, we compared the accuracy of effect size estimates between single-laboratory and multi-laboratory study designs. Single-laboratory studies generally failed to predict effect size accurately, and larger sample sizes rendered effect size estimates even less accurate. By contrast, multi-laboratory designs including as few as 2 to 4 laboratories increased coverage probability by up to 42 percentage points without a need for larger sample sizes. These findings demonstrate that within-study standardization is a major cause of poor reproducibility. More representative study samples are required to improve the external validity and reproducibility of preclinical animal research and to prevent wasting animals and resources for inconclusive research.

Upwash exploitation and downwash avoidance by flap phasing in ibis formation flight.

Many species travel in highly organized groups. The most quoted function of these configurations is to reduce energy expenditure and enhance locomotor performance of individuals in the assemblage. The distinctive V formation of bird flocks has long intrigued researchers and continues to attract both scientific and popular attention. The well-held belief is that such aggregations give an energetic benefit for those birds that are flying behind and to one side of another bird through using the regions of upwash generated by the wings of the preceding bird, although a definitive account of the aerodynamic implications of these formations has remained elusive. Here we show that individuals of northern bald ibises (Geronticus eremita) flying in a V flock position themselves in aerodynamically optimum positions, in that they agree with theoretical aerodynamic predictions. Furthermore, we demonstrate that birds show wingtip path coherence when flying in V positions, flapping spatially in phase and thus enabling upwash capture to be maximized throughout the entire flap cycle. In contrast, when birds fly immediately behind another bird--in a streamwise position--there is no wingtip path coherence; the wing-beats are in spatial anti-phase. This could potentially reduce the adverse effects of downwash for the following bird. These aerodynamic accomplishments were previously not thought possible for birds because of the complex flight dynamics and sensory feedback that would be required to perform such a feat. We conclude that the intricate mechanisms involved in V formation flight indicate awareness of the spatial wake structures of nearby flock-mates, and remarkable ability either to sense or predict it. We suggest that birds in V formation have phasing strategies to cope with the dynamic wakes produced by flapping wings.

Dynamic network measures reveal the impact of cattle markets and alpine summering on the risk of epidemic outbreaks in the Swiss cattle population.

BACKGROUND: Livestock herds are interconnected with each other via an intricate network of transports of animals which represents a potential substrate for the spread of epidemic diseases. We analysed four years (2012-2015) of daily bovine transports to assess the risk of disease transmission and identify times and locations where monitoring would be most effective. Specifically, we investigated how the seasonal dynamics of transport networks, driven by the alpine summering and traditional cattle markets, affect the risk of epidemic outbreaks. RESULTS: We found strong and consistent seasonal variation in several structural network measures as well as in measures for outbreak risk. Analysis of the consequences of excluding markets, dealers and alpine pastures from the network shows that markets contribute much more to the overall outbreak risk than alpine summering. Static descriptors of monthly transport networks were poor predictors of outbreak risk emanating from individual holdings; a dynamic measure, which takes the temporal structure of the network into account, gave better risk estimates. A stochastic simulation suggests that targeted surveillance based on this dynamic network allows a higher detection rate and smaller outbreak size at detection than compared to other sampling schemes. CONCLUSIONS: Dynamic measures based on time-stamped data-the outgoing contact chain-can give better risk estimates and could help to improve surveillance schemes. Using this measure we find evidence that even in a country with intense summering practice, markets continue being the prime risk factor for the spread of contagious diseases.

Matching times of leading and following suggest cooperation through direct reciprocity during V-formation flight in ibis.

One conspicuous feature of several larger bird species is their annual migration in V-shaped or echelon formation. When birds are flying in these formations, energy savings can be achieved by using the aerodynamic up-wash produced by the preceding bird. As the leading bird in a formation cannot profit from this up-wash, a social dilemma arises around the question of who is going to fly in front? To investigate how this dilemma is solved, we studied the flight behavior of a flock of juvenile Northern bald ibis (Geronticus eremita) during a human-guided autumn migration. We could show that the amount of time a bird is leading a formation is strongly correlated with the time it can itself profit from flying in the wake of another bird. On the dyadic level, birds match the time they spend in the wake of each other by frequent pairwise switches of the leading position. Taken together, these results suggest that bald ibis cooperate by directly taking turns in leading a formation. On the proximate level, we propose that it is mainly the high number of iterations and the immediacy of reciprocation opportunities that favor direct reciprocation. Finally, we found evidence that the animals' propensity to reciprocate in leading has a substantial influence on the size and cohesion of the flight formations.

Wild birds respond to flockmate loss by increasing their social network associations to others.

Understanding the consequences of losing individuals from wild populations is a current and pressing issue, yet how such loss influences the social behaviour of the remaining animals is largely unexplored. Through combining the automated tracking of winter flocks of over 500 wild great tits (Parus major) with removal experiments, we assessed how individuals' social network positions responded to the loss of their social associates. We found that the extent of flockmate loss that individuals experienced correlated positively with subsequent increases in the number of their social associations, the average strength of their bonds and their overall connectedness within the social network (defined as summed edge weights). Increased social connectivity was not driven by general disturbance or changes in foraging behaviour, but by modifications to fine-scale social network connections in response to losing their associates. Therefore, the reduction in social connectedness expected by individual loss may be mitigated by increases in social associations between remaining individuals. Given that these findings demonstrate rapid adjustment of social network associations in response to the loss of previous social ties, future research should examine the generality of the compensatory adjustment of social relations in ways that maintain the structure of social organization.

The evolution of generalized reciprocity in social interaction networks.

Generalized reciprocity has been proposed as a mechanism for enabling continued cooperation between unrelated individuals. It can be described by the simple rule "help somebody if you received help from someone", and as it does not require individual recognition, complex cognition or extended memory capacities, it has the potential to explain cooperation in a large number of organisms. In a panmictic population this mechanism is vulnerable to defection by individuals who readily accept help but do not help themselves. Here, I investigate to what extent the limitation of social interactions to a social neighborhood can lead to conditions that favor generalized reciprocity in the absence of population structuring. It can be shown that cooperation is likely to evolve if one assumes certain sparse interaction graphs, if strategies are discrete, and if spontaneous helping and reciprocating are independently inherited.

Heterogeneity of animal experiments and how to deal with it.

Heterogeneity of study samples is ubiquitous in animal experiments. Here, we discuss the different options of how to deal with heterogeneity in the statistical analysis of a single experiment. Specifically, data from different sub-groups (e.g. sex, strain, age cohorts) may be analysed separately, heterogenization factors may be ignored and data pooled for analysis, or heterogenization factors may be included as additional variables in the statistical model. The cost of ignoring a heterogenization factor is an inflated estimate of the variance and a consequent loss of statistical power. Therefore, it is usually preferable to include the heterogenization factor in the statistical model, especially if the heterogenization factor has been introduced intentionally (e.g. using both sexes). If heterogenization factors are included, they can be treated either as fixed factors in an analysis of variance design or sometimes as random effects in mixed effects regression models. Finally, for an appropriate sample size estimation, it is necessary to decide whether to treat heterogenization factors as nuisance variables, or whether the experiment should be powered to be able to detect not only the main effect of the treatment but also interactions between heterogenization factors and the treatment variable.

Evidence of HARKing in mouse behavioural tests of anxiety.

Over the last decades, behavioural tests in animals, especially rodents, have been a standard screening method to determine the mechanisms of action and efficacy of psychopharmacological compounds. Yet, recently the reproducibility of some of these tests has been questioned. Based on a systematic review of the sensitivity of mouse behavioural tests to anxiolytic drugs, we analysed behavioural outcomes extracted from 206 studies testing the effect of diazepam in either the open-field test or the hole-board test. Surprisingly, we found that both the rationale given for using the test, whether to detect anxiolytic or sedative effects, and the predicted effect of diazepam, anxiolytic or sedative, strongly depended on the reported test results. The most likely explanation for such strong dependency is post hoc reasoning, also called hypothesizing after the results are known (HARKing). HARKing can invalidate study outcomes and hampers evidence synthesis by inflating effect sizes. It may also lead researchers into blind alleys, and waste animals, time and resources for inconclusive research.

The impact of cage dividers on mouse aggression, dominance and hormone levels.

Home cage aggression in group-housed male mice is a major welfare concern and may compromise animal research. Conventional cages prevent flight or retreat from sight, increasing the risk that agonistic encounters will result in injury. Moreover, depending on social rank, mice vary in their phenotype, and these effects seem highly variable and dependent on the social context. Interventions that reduce aggression, therefore, may reduce not only injuries and stress, but also variability between cage mates. Here we housed male mice (Balb/c and SWISS, group sizes of three and five) with or without partial cage dividers for two months. Mice were inspected for wounding weekly and home cages were recorded during housing and after 6h isolation housing, to assess aggression and assign individual social ranks. Fecal boli and fur were collected to quantify steroid levels. We found no evidence that the provision of cage dividers improves the welfare of group housed male mice; The prevalence of injuries and steroid levels was similar between the two housing conditions and aggression was reduced only in Balb/c strain. However, mice housed with cage dividers developed less despotic hierarchies and had more stable social ranks. We also found a relationship between hormone levels and social rank depending on housing type. Therefore, addition of cage dividers may play a role in stabilizing social ranks and modulating the activation of hypothalamic-pituitary-adrenal (HPA) and hypothalamic-pituitary-gonadal (HPG) axes, thus reducing phenotypic variability between mice of different ranks.

Differential effects of purified low molecular weight Poly(I:C) in the maternal immune activation model depend on the laboratory environment.

The Poly (I:C) (polyriboinosinic-polyribocytidilic acid) paradigm of maternal immune activation (MIA) is most widely used as experimental model for the evaluation of the effects of gestational infection on the brain and behavior of the progeny. We have previously reported significant batch-to-batch variability in the effects of Poly (I:C), purchased from the same supplier (Sigma-Aldrich), on maternal and fetal immune responses and found these differences to be dependent on the relative amount of synthetic double-stranded RNA fragments in the high versus low molecular weight (LMW) range contained in the compound. We here resorted to Poly (I:C) purified for LMW dsRNA fragments to establish a MIA paradigm with increased reproducibility and enhanced standardization in an effort to refine the MIA paradigm and characterize its effect on offspring behavior. We found that the parallel application of LMW Poly (I:C) in two different MIA-experienced laboratories (Vienna and Zurich) yielded differential outcomes in terms of maternal immune responses and behavioral phenotypes in the offspring generation. In both experimental sites, administration of LMW Poly (I:C) induced a significant sickness response and cytokine induction in the pregnant dam and fetal brains, while the expected deficit in sociability as one main behavioral outcome parameter in the MIA progeny, was only present in the Zurich, but not the Vienna cohort. We conclude that although using Poly (I:C) purified for a defined molecular weight range reduces batch-to-batch variability, it does not make the MIA model more reliable and robust. The differential response in behavioral phenotypes of the MIA offspring between the two laboratories illustrates the highly complex interaction between prenatal and postnatal milieus - including the laboratory environment - that determine offspring phenotypic outcomes after MIA. Consequently, establishing a new MIA protocol or implementing the MIA model firstly under new or changed environmental conditions must include the assessment of offspring behavior to ensure solid and reproducible experimental outcomes.

Using mice from different breeding sites fails to improve replicability of results from single-laboratory studies.

Theoretical and empirical evidence indicates that low external validity due to rigorous standardization of study populations is a cause of poor replicability in animal research. Here we report a multi-laboratory study aimed at investigating whether heterogenization of study populations by using animals from different breeding sites increases the replicability of results from single-laboratory studies. We used male C57BL/6J mice from six different breeding sites to test a standardized against a heterogenized (HET) study design in six independent replicate test laboratories. For the standardized design, each laboratory ordered mice from a single breeding site (each laboratory from a different one), while for the HET design, each laboratory ordered proportionate numbers of mice from the five remaining breeding sites. To test our hypothesis, we assessed 14 outcome variables, including body weight, behavioral measures obtained from a single session on an elevated plus maze, and clinical blood parameters. Both breeding site and test laboratory affected variation in outcome variables, but the effect of test laboratory was more pronounced for most outcome variables. Moreover, heterogenization of study populations by breeding site (HET) did not reduce variation in outcome variables between test laboratories, which was most likely due to the fact that breeding site had only little effect on variation in outcome variables, thereby limiting the scope for HET to reduce between-lab variation. We conclude that heterogenization of study populations by breeding site has limited capacity for improving the replicability of results from single-laboratory animal studies.

Characterization of bird formations using fuzzy modelling.

The investigation of the emergent collective behaviour in flying birds is a challenging task, yet it has always fascinated scientists from different disciplines. In the attempt of studying and modelling line formation, we collected high-precision position data of 29 free-flying northern bald ibises (Geronticus eremita) using Global Navigation Satellite System loggers, to investigate whether the spatial relationships within a flock can be explained by birds maintaining energetically advantageous positions. Specifically, we exploited domain knowledge and available literature information to model by means of fuzzy logic where the air vortices lie behind a flying bird. This allowed us to determine when a leading bird provides the upwash to a following bird, reducing its overall effort. Our results show that the fuzzy model allows to easily distinguish which bird is flying in the wake of another individual, provides a clear indication about flying flock dynamics and also gives a hint about birds' social relationships.