Biology subjects

Ostner, J.

Publications and source records attributed to Ostner, J..

At least 19 recordsLinked to original sources

bioRxiv2025

Inferring the causes of animal social network structure from time-series data

Behavioural ecologists aim to understand the causes of animal social structure. Connecting theoretical models of social structure with empirical observations remains, however, a for-midable challenge. While most of the current statistical methods for animal social network analysis rely on data that are aggregated over time and summarised as one behavioural dimension (e.g., an adjacency-matrix), common behavioural sampling techniques (e.g., focal-animal sampling) produce data in continuous time, and involve different behaviours. Furthermore, empiricists in the field are generally interested in causal inference, but lack a framework to rigorously analyse focal-animal sampling data in light of transparent causal assumptions. As a consequence, common methods are often inappropriate, and can lead to wrong biological conclusions. Here, we introduce a causal Bayesian modelling framework to empirically study the causes of social network structure from focal-animal sampling data. We start by outlining a generative model that encodes how biological and measurement processes jointly produce social network data in continuous time; namely, as a temporal sequence of dyadic behavioural states (e.g., no body contact, social resting, grooming). Building upon the generative model, we develop a statistical model: a multilevel, multiplex Bayesian model that takes raw focal observations as input, and produces a posterior probability distribution for the generative parameters as output. After validating the statistical models performance with sparse data-- common in real-world settings--we illustrate its application with an empirical data set collected in wild Assamese macaques. We notably showcase how researchers can compute probabilistic estimates for well-defined causal hypotheses about the drivers of social structure. With this work, we not only contribute novel theoretical and statistical tools to the field, but also illustrate a workflow that allows researchers to iteratively translate their domain expertise into a formal analytical strategy--bridging theoretical and empirical research in behavioural ecology.

animal behavior and cognition↗

bioRxiv2025

Complex modulation of visual attention to 3rd person interactions in wild macaques

Social animals rely on socio-cognitive skills to monitor their social environment and make informed decisions. Yet, visual attention to others interactions in real-life scenes remains understudied, despite evidence for dedicated neuronal networks for the processing of real-life social information and for social interactions in particular. Here, we ask how subject characteristics, interaction valence, and social relationships between subject and stimuli interact to guide overt attention of 56 wild male Assamese macaques to 962 3rd person social scenes that unfolded spontaneously in their vicinity. Social interactions drew more and longer attention than social control scenes lacking the interaction, highlighting the relevance of interaction. The effect of both social bond strength and dominance relations between subject and stimulus on the probability to attend was independently modulated by scene valence. Moreover, beyond reacting to the affordances of the social scene and independent of their social status, males differed in how long they observed the interactions of others. These findings suggest that visual attention is guided by the value an individual associates with different aspects of the social scene based on previous experience in both the agonistic and the affiliative realms and by current threat indicative of complex information integration in the brain.

animal behavior and cognition↗

bioRxiv2025

Spontaneous eye blinking is influenced by activity, synchronized with attentional breakpoints, but not modulated by social relevance in Barbary macaques

Spontaneous eye blinking is a ubiquitous behavior in animals including humans necessary for lubricating the ocular surface and preventing dryness. Beyond this functional role, eye blinking also provides a window into an animals cognitive state and attention allocation. Here, we investigated in 13 female Barbary macaques the modulation of spontaneous eye blinking during two naturally occurring activities differing in their attentional demand (resting and allo-grooming) and additionally assess the influence of social relevance of the interaction on attention allocation. Eye blink rates were significantly lower during grooming compared to resting, suggesting increased attention during this cognitively more demanding task. Dominance rank difference and affiliative relationship strength between the groomer and groomee did not additionally influence eye blink rate. Blinking was timed to coincide with ingestion events during grooming, which may serve as explicit attentional breakpoints. By systematically timing blinks with periods of decreased visual demand, macaques effectively minimized information loss during the non-visual phase of the grooming process. Our study provides insights into the regulation of spontaneous eye blinking in nonhuman animals using a non-invasive tool for the study of visual attention and cognitive load.

animal behavior and cognition↗

bioRxiv2025

Age-trajectory of mother-infant relationships in wild Assamese macaques

Maternal care is ubiquitous in mammals, yet its degree and duration vary across taxa. Primate mothers provide extended care, with similar developmental transitions of the mother-infant relationship, though with different paces of change. Ecological conditions can influence the trajectory of this relationship, but data from the wild are still scarce. We used methods from growth studies to quantitatively describe the non-linear age-trajectory of the mother-infant spatial relationship, and the transition from dependent to independent feeding and locomotion in wild Assamese macaques (M. assamensis). We also explored sex differences in the development of the mother-infant relationship. We used a modified Gompertz function to model the combined effect of infant age and sex on mother and infant behaviors extracted from focal observations of 58 infants. Newborns were fully dependent on their mothers for feeding and transportation, with mothers maintaining close proximity. A transitional phase emerged between 1 and 3 months of infant age, marked by a noticeable reduction in the spatial proximity with the mother and a shift in the responsibility for the infants feeding and transportation. During the second half of infancy, the decrease in proximity time slowed down, with infants achieving near-complete locomotion independence, spending the majority of time away from their mothers and feeding independently. No sex differences were found. Our models provided a robust fit for most variables, but we recommend future exploration of alternative nonlinear functions. We interpret the early infant independence observed in our population in the context of the species reproductive strategy.

animal behavior and cognition↗

bioRxiv2024

Score matching for differential abundance testing of compositional high-throughput sequencing data

The class of a-b power interaction models, proposed by Yu et al. (2024), provides a general framework for modeling sparse compositional count data with pairwise feature interactions. This class includes many distributions as special cases and enables zero count handling through power transformations, making it especially suitable for modern high-throughput sequencing data with excess zeros, including single-cell RNA-Seq and amplicon sequencing data. Here, we present an extension of this class of models that can include covariate information, allowing for accurate characterization of covariate dependencies in heterogeneous populations. Combining this model with a tailored differential abundance (DA) test leads to a novel DA testing scheme, cosmoDA, that can reduce false positive detection caused by correlated features. cosmoDA uses the generalized score matching estimation framework for power interaction models Our benchmarks on simulated and real data show that cosmoDA can accurately estimate feature interactions in the presence of population heterogeneity and significantly reduces the false discovery rate when testing for differential abundance of correlated features. Finally, cosmoDA provides an explicit link to popular Box-Cox-type data transformations and allows to assess the impact of zero replacement and power transformations on downstream differential abundance results. cosmoDA is available at https://github.com/bio-datascience/cosmoDA.

bioinformatics↗

bioRxiv2024

PriMAT: A robust multi-animal tracking model for primates in the wild

O_LIDetection and tracking of animals is an important first step for automated behavioral studies using videos. Animal tracking is currently done mostly using deep learning frameworks based on keypoints, which show remarkable results in lab settings with fixed cameras, backgrounds, and lighting. However, multi-animal tracking in the wild presents several challenges such as high variability in background and lighting conditions, complex motion, and occlusion. C_LIO_LIWe propose a multi-animal tracking model, PriMAT, for nonhuman primates in the wild. The model learns to detect and track primates and other objects of interest from labeled videos or single images using bounding boxes instead of keypoints. Using bounding boxes significantly facilitates data annotation and robustness. Our one-stage model is conceptually simple but highly flexible, and we add a classification branch that allows us to train individual identification. C_LIO_LITo evaluate the performance of our model, we applied it in two case studies with Assamese macaques (Macaca assamensis) and redfronted lemurs (Eulemur rufifrons) in the wild. We show that with only a few hundred frames labeled with bounding boxes, we can achieve robust tracking results. Combining these results with the classification branch for the lemur videos, our model shows an accuracy of 84% in predicting lemur identities. C_LIO_LIOur approach presents a promising solution for accurately tracking and identifying animals in the wild, offering researchers a tool to study animal behavior in their natural habitats. Our code, models, training images, and evaluation video sequences are publicly available1, facilitating their use for animal behavior analyses and future research in this field. C_LI

animal behavior and cognition↗

bioRxiv2024

Pertpy: an end-to-end framework for perturbation analysis

Advances in single-cell technology have enabled the measurement of cell-resolved molecular states across a variety of cell lines and tissues under a plethora of genetic, chemical, environmental, or disease perturbations. Current methods focus on differential comparison or are specific to a particular task in a multi-condition setting with purely statistical perspectives. The quickly growing number, size, and complexity of such studies requires a scalable analysis framework that takes existing biological context into account. Here, we present pertpy, a Python-based modular framework for the analysis of large-scale perturbation single-cell experiments. Pertpy provides access to harmonized perturbation datasets and metadata databases along with numerous fast and user-friendly implementations of both established and novel methods such as automatic metadata annotation or perturbation distances to efficiently analyze perturbation data. As part of the scverse ecosystem, pertpy interoperates with existing libraries for the analysis of single-cell data and is designed to be easily extended.

bioinformatics↗

bioRxiv2024

BacSC: A general workflow for bacterial single-cell RNA sequencing data analysis

Bacterial single-cell RNA sequencing has the potential to elucidate within-population heterogeneity of prokaryotes, as well as their interaction with host systems. Despite conceptual similarities, the statistical properties of bacterial single-cell datasets are highly dependent on the protocol, making proper processing essential to tap their full potential. We present BacSC, a fully data-driven computational pipeline that processes bacterial single-cell data without requiring manual intervention. BacSC performs data-adaptive quality control and variance stabilization, selects suitable parameters for dimension reduction, neighborhood embedding, and clustering, and provides false discovery rate control in differential gene expression testing. We validated BacSC on a broad selection of bacterial single-cell datasets spanning multiple protocols and species. Here, BacSC detected subpopulations in Klebsiella pneumoniae, found matching structures of Pseudomonas aeruginosa under regular and low-iron conditions, and better represented subpopulation dynamics of Bacillus subtilis. BacSC thus simplifies statistical processing of bacterial single-cell data and reduces the danger of incorrect processing.

bioinformatics↗

bioRxiv2024

A causal framework for the drivers of animal social network structure

AO_SCPLOWBSTRACTC_SCPLOWA major goal of behavioural ecology is to explain how phenotypic and ecological factors shape the networks of social relationships that animals form with one another. This inferential task is notoriously challenging. The social networks of interest are generally not observed, but must be approximated from behavioural samples. Moreover, these data are highly dependent: the observed network edges correlate with one another, due to biological and sampling processes. Failing to account for the resulting uncertainty and biases can lead to dysfunctional statistical procedures, and thus to incorrect results. Here, we argue that these problems should be understood--and addressed--as problems of causal inference. For this purpose, we introduce a Bayesian causal modelling framework that explicitly defines the links between the target interaction network, its causes, and the data. We illustrate the mechanics of our framework with simulation studies and an empirical example. First, we encode causal effects of individual-, dyad-, and group-level features on social interactions using Directed Acyclic Graphs and Structural Causal Models. These quantities are the objects of inquiry, our estimands. Second, we develop estimators for these effects--namely, Bayesian multilevel extensions of the Social Relations Model. Third, we recover the structural parameters of interest, map statistical estimates to the underlying causal structures, and compute causal estimates from the joint posterior distribution. Throughout the manuscript, we develop models layer by layer, thereby illustrating an iterative workflow for causal inference in social networks. We conclude by summarising this workflow as a set of seven steps, and provide practical recommendations.

animal behavior and cognition↗

bioRxiv2024

MetaIBS: large-scale amplicon-based meta analysis of irritable bowel syndrome

BackgroundIrritable Bowel Syndrome (IBS) is a chronic functional bowel disorder causing abdominal discomfort, as well as transit deregulation with constipation and/or diarrhea. The pathophysiology of IBS is poorly understood and believed to be multifactorial. The role of gut microbiota in IBS has been investigated in several case-control studies, in particular via 16S rRNA amplicon sequencing surveys. These studies, however, have not yet led to a consistent picture of significant changes in gut microbial compositions across health and disease. One key bottleneck is the modest cohort sizes of most individual studies and a high diversity of experimental, bioinformatics, and statistical analysis approaches across studies. ResultsWe address these shortcomings by presenting MetaIBS, an open-access data repository and associated meta-analysis workflow of thirteen 16S rRNA amplicon datasets comprising both fecal matter and sigmoid biopsy samples spanning {bsim}2,500 IBS and healthy individuals. MetaIBS includes a tailored computational framework that (i) enables coherent de novo processing and taxonomic assignments of the raw 16S rRNA amplicon reads across experimental protocols and sequencing technologies, and (ii) statistical workflows for visualization and analysis at different taxonomic ranks and data granularity. Our statistical meta-analysis shows that popular high-level microbiome summary statistics, including Firmicutes/Bacteroidota ratios or diversity indices, are insufficient for reliable discrimination between IBS patients and healthy controls. Fine-grained multi-method differential abundance and classification analysis, however, can identify sets of differentially abundant taxa that replicate across multiple datasets, including Coprococcus eutactus and Alistipes finegoldii. ConclusionsMetaIBS provides a curated and reproducible data and (meta-)analysis resource for amplicon-based IBS research at unprecedented scale. MetaIBS allows assessing the heterogeneity of IBS cohorts across multiple experimental protocols, sample types, and IBS phenotypes. Our framework will likely contribute to more coherent insights into the role of the microbiome in IBS and the discovery of reliable microbial IBS biomarkers for follow-up functional and translational studies.

microbiology↗

bioRxiv2023

MacaqueNet: big-team research into the biological drivers of social relationships

O_LIFor many animals, social relationships are a key determinant of fitness. However, major gaps remain in our understanding of the adaptive function, ontogeny, evolution, and mechanistic underpinnings of social relationships. There is a vast and ever-accumulating amount of social behavioural data on individually recognised animals, an incredible resource to shed light onto the biological basis of social relationships. Yet, the full potential of such data lies in comparative research across taxa with distinct life histories and ecologies. Substantial challenges impede systematic comparisons, one of which is the lack of persistent, accessible, and standardised databases. C_LIO_LIHere, we advocate for the creation of big-team collaborations and comparative databases to unlock the wealth of behavioural data for research on social relationships by introducing MacaqueNet (https://macaquenet.github.io/). C_LIO_LIAs a global collaboration of over 100 researchers, the MacaqueNet database encompasses data from 1981 to the present on 14 species and is the first publicly searchable and standardised database on affiliative and agonistic animal social networks. With substantial inter-specific variation in ecology and social structure and the first published record on macaque behaviour dating back to 1956, macaque research has already contributed to answering fundamental questions on the biological bases and evolution of social relationships. Building on these strong foundations, we believe that MacaqueNet can further promote collaborative and comparative research on social behaviour. C_LIO_LIWe believe that big-team approaches to building standardised databases, bringing together data contributors and researchers, will aid much-needed large-scale comparative research in behavioural ecology and beyond. We describe the establishment of MacaqueNet, from starting a large-scale collective to the creation of a cross-species collaborative database and the implementation of data entry and retrieval protocols. As such, we hope to provide a functional example for future endeavours of large-scale collaborative research into the biology of social behaviour. C_LI

animal behavior and cognition↗

bioRxiv2021

tascCODA: Bayesian tree-aggregated analysis of compositional amplicon and single-cell data

Accurate generative statistical modeling of count data is of critical relevance for the analysis of biological datasets from high-throughput sequencing technologies. Important instances include the modeling of microbiome compositions from amplicon sequencing surveys and the analysis of cell type compositions derived from single-cell RNA sequencing. Microbial and cell type abundance data share remarkably similar statistical features, including their inherent compositionality and a natural hierarchical ordering of the individual components from taxonomic or cell lineage tree information, respectively. To this end, we introduce a Bayesian model for tree-aggregated amplicon and single-cell compositional data analysis (tascCODA) that seamlessly integrates hierarchical information and experimental covariate data into the generative modeling of compositional count data. By combining latent parameters based on the tree structure with spike-and-slab Lasso penalization, tascCODA can determine covariate effects across different levels of the population hierarchy in a data-driven parsimonious way. In the context of differential abundance testing, we validate tascCODAs excellent performance on a comprehensive set of synthetic benchmark scenarios. Our analyses on human single-cell RNA-seq data from ulcerative colitis patients and amplicon data from patients with irritable bowel syndrome, respectively, identified aggregated cell type and taxon compositional changes that were more predictive and parsimonious than those proposed by other schemes. We posit that tascCODA1 constitutes a valuable addition to the growing statistical toolbox for generative modeling and analysis of compositional changes in microbial or cell population data.

bioinformatics↗

bioRxiv2020

scCODA: A Bayesian model for compositional single-cell data analysis

Compositional changes of cell types are main drivers of biological processes. Their detection through single-cell experiments is difficult due to the compositionality of the data and low sample sizes. We introduce scCODA (https://github.com/theislab/scCODA), a Bayesian model addressing these issues enabling the study of complex cell type effects in disease, and other stimuli. scCODA demonstrated excellent detection performance and identified experimentally verified cell type changes that were missed in original analyses.

bioinformatics↗

bioRxiv2020

Grooming interventions in female rhesus macaques as social niche construction

Social animals invest time and resources into building and adapting their social environment, which emerges not only from their own but also from the decisions of other group members. Thus, individuals have to monitor interactions between others and potentially decide when and how to interfere to prevent damage to their own investment. These interventions can be subtle, as in the case of affiliative interactions such as grooming, but they can inform us about how animals structure their world and influence other group members. Here, we used interventions into grooming bouts in 29 female rhesus macaques (Macaca mulatta) to determine who intervened into which grooming bouts, why, and what determined intervention outcomes, based on kinship, dominance rank, and affiliative relationships between groomers and (potential) interveners. Using 1132 grooming bouts and 521 interventions, we show that high dominance rank of groomers reduced the risk of intervention. Bystanders, particularly when high-ranking, intervened in grooming of their kin, close affiliates, and close-ranked competitors. Interveners gained access to their close affiliates for subsequent grooming. Affiliative relationship and rank determined intervention outcomes, with reduced aggression risk facilitating grooming involving three individuals. Thus, interventions in this species involved the monitoring of grooming interactions, decision-making based on several individual and dyadic characteristics, and potentially allowed individuals to broaden their access to grooming partners, protect their own relationships, and influence their social niche.

animal behavior and cognition↗

bioRxiv2019

Testosterone and cortisol are negatively associated with ritualized bonding behavior in male macaques

Neuroendocrine research on the formation of social bonds has primarily focused on the role of nonapeptides. However, steroid hormones often act simultaneously to either inhibit or facilitate bonding. Testosterone is proposed to mediate a trade-off between male mating effort and nurturing behavior; therefore, low levels are predicted during periods of nurturing infant care and social bonding. In species where social bonding and support regulates hypothalamic-pituitary-adrenal axis activity, we also expect glucocorticoid levels to be low during bonding periods. We investigated how immunoreactive urinary testosterone (iuT) and cortisol (iuC) were related to triadic male-infant-male interactions - a ritualized male bonding behavior - as well as infant care in male Barbary macaques (Macaca sylvanus). We collected >3000 hours of behavioral observation data during full-day focal animal follows from 14 adult males and quantified iuT and iuC from 650 urine samples. As predicted, both iuT and iuC were negatively correlated with rates of triadic interactions within-subjects in the hours preceding urination. We found no relationship between iuT and iuC with triadic interactions between-subjects. Infant care was weakly positively correlated to iuT and iuC within-subjects, but not between-subjects. The observed negative relationship between iuT and triadic interactions may be beneficial to lower competitive tendencies between adult males and to not inhibit bond formation. Lowered iuC could reflect increased bonding and perceived social support as triadic interactions predict future coalition formation in this species. Additionally, lowered iuC may be reflective of buffered tensions between males. The positive relationship of iuT and iuC with infant care suggests that the handling of infants in may be less nurturing but rather protective or competitive in this species. Measuring steroid hormones in relation to bonding and nurturing can help us interpret behaviors within the ecological contexts that they occur.

animal behavior and cognition↗

bioRxiv2019

Female Assamese macaques bias their affiliation to paternal and maternal kin

Forming strong social bonds leads to higher reproductive success, increased longevity and/or increased infant survival in several mammal species. Given these adaptive benefits, understanding what determines partner preferences in social bonding is important. Maternal relatedness strongly predicts partner preference across many mammalian taxa. Although paternal and maternal kin share the same number of genes, and theoretically similar preferences would therefore be expected for paternal kin, the role of paternal relatedness has received relatively little attention. Here, we investigate the role of maternal and paternal relatedness for female bonding in Assamese macaques (Macaca assamensis), a species characterized by a relatively low male reproductive skew. We studied a wild population under natural conditions using extensive behavioural data and relatedness analyses based on pedigree reconstruction. We found stronger social bonds and more time spent grooming between maternal kin and paternal half-sisters compared to non-kin, with no preference of maternal over paternal kin. Paternally related and non-related dyads did not form stronger bonds when they had less close maternal kin available, however we would need a bigger sample size to confirm this. As expected given the low reproductive skew, bonds between paternal half-sisters closer in age were not stronger than between paternal half-sisters with larger age differences, suggesting that age similarity was not the mechanism by which paternally related individuals recognized each other. An alternative way through which paternal kin could get familiarized is mother- and/or father-mediated familiarity.

animal behavior and cognition↗

bioRxiv2019

Oxytocin increases after affiliative interactions in male Barbary macaques

Mammals living in stable social groups often mitigate the costs of group living through the formation of social bonds and cooperative relationships. The neuropeptide hormone oxytocin (OT) has been proposed to promote both bonding and cooperation although only a limited number of studies have investigated this under natural conditions. Our aim was to assess the role of OT in bonding and cooperation in male Barbary macaques (Macaca sylvanus). First we tested for an effect of affiliation - grooming and triadic male-infant-male interactions - with bond and non-bond partners on urinary OT levels. Secondly we aimed to test whether grooming interactions (and thus increased OT levels) increase a males general propensity to cooperate in polyadic conflicts. We collected behavioral data via full-day focal animal protocols on 14 adult males and measured endogenous OT levels from 139 urine samples collected after affiliation and non-social control periods. Urinary OT levels were higher after grooming with any partner. By contrast, OT levels after male-infant-male interactions with any partner or with bond partners were not different from controls but were higher after interactions with non-bond partners. Previous grooming did not increase the likelihood of males to support others in conflicts. Collectively, our results support research indicating that OT is involved in the regulation of adult social bonds, including in non-reproductive contexts. However, our male-infant-male interaction results go against previous studies suggesting that it is affiliation with bond rather than non-bond partners that trigger the release of OT. Alternatively, OT levels may have been elevated prior to male-infant-male interactions thus facilitating interaction between non-bond partners. The lack of an association of grooming (and by extension increased OT levels) and subsequent support speaks against an OT linked increase in the general propensity to cooperate, yet further studies are needed for a more direct test including the possibility of partner-specific contingent cooperation.

animal behavior and cognition↗

bioRxiv2019

Personality homophily affects male social bonding in wild Assamese macaques (Macaca assamensis)

Animal social bonds are defined as stable, equitable and strong affiliative and cooperative relationships similar to human friendships. Just as human friendships, social bonds are thought to function as alliances that generate adaptive benefits via support in critical situations. In humans, similarity in many sociodemographic, behavioural and intrapersonal characteristics leads to trust and is predictive of friendships. Specifically, personality homophily, that is the tendency of individuals to form social bonds with others who have a similar personality, may increase predictability and facilitate trust and reciprocity among partners with compatible behavioural tendencies. While evidence for social bonding in nonhumans is accumulating, far less is known about its predictors. Here, personality homophily effects on the formation and maintenance of social bonds are shown in twenty-four wild male Assamese macaques (Macaca assamensis), at Phu Khieo Wildlife Sanctuary, Thailand. Dyadic bond strength increased with increased similarity in the trait Connectedness (i.e. frequent and diverse neighbours in 5m proximity and pronounced social tolerance, as high rates of friendly approaches to and by others). To differentiate whether homophily indeed predicted bond formation or whether bonded males personalities became more similar over time, we tested the stability of the connectedness traits in a subset of immigrating males that had to form new bonds. Connectedness in these males remained stable suggesting that males do not adapt their personality to their partner. Our results support the idea of a shared evolutionary origin of homophily as a partner choice strategy in human and non-human animals. The main selective advantage of personality similarity in animal social bonds may result from a more reliable cooperation among individuals with similar cooperative behavioural tendencies.

animal behavior and cognition↗