Search

Peptides are ubiquitous and important biologically derived molecules, that have been found to self-assemble to form a wide array of structures. Extensive research has explored the impacts of both internal chemical composition and external environmental stimuli on the self-assembly behaviour of these systems. However, there is yet to be a systematic study that gathers this rich literature data and collectively examines these experimental factors to provide a global picture of the fundamental rules that govern protein self-assembly behavior. In this work, we curate a peptide assembly database through a combination of manual processing by human experts and literature mining facilitated by a large language model. As a result, we collect more than 1,000 experimental data entries with information about peptide sequence, experimental conditions and corresponding self-assembly phases. Utilizing the collected data, ML models are trained and evaluated, demonstrating excellent accuracy (>80\%) and efficiency in peptide assembly phase classification. Moreover, we fine-tune our GPT model for peptide literature mining with the developed dataset, which exhibits markedly superior performance in extracting information from academic publications relative to the pre-trained model. We find that this workflow can substantially improve efficiency when exploring potential self-assembling peptide candidates, through guiding experimental work, while also deepening our understanding of the mechanisms governing peptide self-assembly. In doing so, novel structures can be accessed for a range of applications including sensing, catalysis and biomaterials.

A new disturbance observer based control scheme is developed for a quadrotor under the concurrent disturbances from a lightweight elastic tether cable and a lumped vertical disturbance. This elastic tether is unusual as it creates a disturbance proportional to the multicopter's translational movement. This paper takes an observer-based approach to estimate the stiffness coefficient of the cable and uses the system model to update the estimates of the external forces, which are then compensated in the control action. Given that the tethered cable force affects both horizontal channels of the quadrotor and is also coupled with the vertical channel, the proposed disturbance observer is constructed to exploit the redundant measurements across all three channels to jointly estimate the cable stiffness and the vertical disturbance. A pseudo-inverse method is used to determine the observer gain functions, such that the estimation of the two quantities is decoupled and stable. Compared to standard disturbance observers which assume nearly constant disturbances, the proposed approach can quickly adjust its total force estimate as the tethered quadrotor changes its position or tautness of the tether. This is applied to two experiments - a tracking performance test where the multicopter moves under a constant tether strain, and an object extraction test. In the second test, the multicopter manipulates a nonlinear mechanism mimicking the extraction of a wedged object. In both cases, the proposed approach shows significant improvement over standard Disturbance Observer and Extended State Observer approaches. A video summary of the experiments can be found at https://youtu.be/9gKr13WTj-k.

The success of machine learning models relies heavily on effectively representing high-dimensional data. However, ensuring data representations capture human-understandable concepts remains difficult, often requiring the incorporation of prior knowledge and decomposition of data into multiple subspaces. Traditional linear methods fall short in modeling more than one space, while more expressive deep learning approaches lack interpretability. Here, we introduce Supervised Independent Subspace Principal Component Analysis ($\texttt{sisPCA}$), a PCA extension designed for multi-subspace learning. Leveraging the Hilbert-Schmidt Independence Criterion (HSIC), $\texttt{sisPCA}$ incorporates supervision and simultaneously ensures subspace disentanglement. We demonstrate $\texttt{sisPCA}$'s connections with autoencoders and regularized linear regression and showcase its ability to identify and separate hidden data structures through extensive applications, including breast cancer diagnosis from image features, learning aging-associated DNA methylation changes, and single-cell analysis of malaria infection. Our results reveal distinct functional pathways associated with malaria colonization, underscoring the essentiality of explainable representation in high-dimensional data analysis., Comment: 10 pages and 6 figures in the main text; To be published in the Proceedings of the 38th Conference on Neural Information Processing Systems (NeurIPS 2024)

The Core Data Ontology (CDO) and the Informatics Domain Model represent a transformative approach to computational systems, shifting from traditional node-centric designs to a data-centric paradigm. This paper introduces a framework where data is categorized into four modalities: objects, events, concepts, and actions. This quadrimodal structure enhances data security, semantic interoperability, and scalability across distributed data ecosystems. The CDO offers a comprehensive ontology that supports AI development, role-based access control, and multimodal data management. By focusing on the intrinsic value of data, the Informatics Domain Model redefines system architectures to prioritize data security, provenance, and auditability, addressing vulnerabilities in current models. The paper outlines the methodology for developing the CDO, explores its practical applications in fields such as AI, robotics, and legal compliance, and discusses future directions for scalable, decentralized, and interoperable data ecosystems., Comment: Proceedings of the 9th International Conference on Data Mining & Knowledge Management (DaKM 2024)

Loop arrangements and their quantum superpositions describe several interesting many-particle states. We propose that they also describe bonding in a class of transition metal dichalcogenides. We present an effective quantum loop model for monolayers with 1T structure and a d$^2$ valence electron configuration: materials of the form MX$_2$ (M = Mo, W and X=S, Se, Te) and AM$'$Y$_2$ (A = Li, Na; M$'$ = V, Nb and Y = O, S, Se). Their t$_{2g}$ orbitals exhibit strongly directional overlaps between neighbouring atoms, favouring the formation of valence bonds. A transition metal atom forms two valence bonds, each with one of its neighbours. When connected, these bonds form loops that cover the triangular lattice. We construct a minimal Rokhsar-Kivelson-like model with resonance processes that cut and reconnect loops that run in proximity. The resulting dynamics is more constrained than in traditional quantum dimer models, with a `bending' constraint that arises from orbital structure. In the resulting phase diagram, we find phases that resemble distorted phases seen in materials, viz., the 1T$'$ and trimerized phases. As a testable prediction, we propose that a single d$^1$ or d$^3$ impurity will terminate a loop and give rise to a long-ranged texture. For example, a Ti/Cr defect in LiVO$_2$ will produce one or more domain walls that propagate outward from the impurity. We discuss the possibility of a loop liquid phase that can emerge in these materials., Comment: 5 pages, 4 figures, includes 3-page supplement

We present the second data release of the MeerKAT Absorption Line Survey (MALS), consisting of wideband continuum catalogues of 391 pointings observed at L~band. The full wideband catalogue covers 4344 deg$^2$ of sky, reaches a depth of 10 $\mu$Jy beam$^{-1}$, and contains 971,980 sources. With its balance between survey depth and sky coverage, MALS DR2 covers five orders of magnitude of flux density, presenting a robust view of the extragalactic radio source population down to 200 $\mu$Jy. Using this catalogue, we perform a measurement of the cosmic radio dipole, an anisotropy in the number counts of radio sources with respect to the cosmic background, analogous to the dipole found in the cosmic microwave background (CMB). For this measurement, we present the characterisation of completeness and noise properties of the catalogue, and show that a declination-dependent systematic affects the number density of faint sources. In the dipole measurement on the MALS catalogue, we recover reasonable dipole measurements once we model the declination systematic with a linear fit between the size of the major axis of the restoring beam and the amount of sources of each pointing. The final results are consistent with the CMB dipole in terms of direction and amplitude, unlike many recent measurements of the cosmic radio dipole made with other centimetre wavelength catalogues, which generally show a significantly larger amplitude. This result demonstrates the value of dipole measurements with deeper and more sparse radio surveys, as the population of faint sources probed may have had a significant impact on the measured dipole., Comment: 25 pages, 14 figures. Accepted for publication in Astronomy & Astrophysics. The MALS wideband catalogues and images are publicly available at https://mals.iucaa.in

Multiwavelength studies of galaxy clusters are crucial for understanding the complex interconnection of the thermal and non-thermal constituents of these massive structures and uncovering the physical processes involved in their formation and evolution. Here, we report a multiwavelength assessment of the galaxy cluster A384, which was previously reported to host a radio halo with a 660 kpc size at MeerKAT 1.28 GHz. The halo is slightly offset from the cluster centre. Our uGMRT observation reveals that the halo extends up to 690 kpc at 407 MHz with a nonuniform spectral index $\alpha^{1284}_{407}$ distribution varying from flat (-0.5) to steep (-1.3) values. In addition, we use legacy GMRT 608 MHz, \textit{XMM-Newton} X-ray, and the Dark Energy Survey optical observations to obtain an extensive understanding of the dynamical nature of the galaxy cluster. The X-ray surface brightness concentration parameter (c$_{SB}$ = 0.16) and centroid shift (w = 0.057) reveal an ongoing dynamical disturbance in the cluster. This is also supported by the elongated 2-D optical galaxy density distribution map of the cluster. We obtain an asymmetry parameter of 0.35 $\pm$ 0.04 from optical analysis, further supporting the dynamical disturbance in the cluster. The radio and X-ray surface brightness follows a sub-linear correlation. Our observation suggests that the cluster is currently in a merging state where particle re-acceleration in the turbulent ICM resulted in the radio halo emission., Comment: 10 pages, 8 Figures, submitted to MNRAS

We present a model to describe the concentration-dependent growth of protein filaments. Our model contains two states, a low entropy/high affinity ordered state and a high entropy/low affinity disordered state. Consistent with experiments, our model shows a diffusion-limited linear growth regime at low concentration, followed by a concentration independent plateau at intermediate concentrations, and rapid disordered precipitation at the highest concentrations. We show that growth in the linear and plateau regions is the result of two processes that compete amid the rapid binding and unbinding of non-specific states. The first process is the addition of ordered molecules during the periods where the end of the filament is free of incorrectly bound molecules. The second process is the capture of defects, which occurs when consecutive ordered additions occur on top of incorrectly bound molecules. We show that a key molecular property is the probability that a diffusive collision results in a correctly bound state. Small values of this probability suppress the defect capture growth mode, resulting in a plateau in the growth rate when incorrectly bound molecules become common enough to poison ordered growth. We show that conditions that non-specifically suppress or enhance intermolecular interactions, such as the addition of depletants or osmolytes, have opposite effects on the growth rate in the linear and plateau regimes. In the linear regime stronger interactions promote growth by reducing dissolution events, but in the plateau regime stronger interactions inhibit growth by stabilizing incorrectly bound molecules., Comment: 20 pages, 7 figures

Many ligands commonly used to prepare nanoparticle catalysts with precise nanoscale features contain nitrogen (e.g., oleylamine); here, we found that the use of nitrogen-containing ligands during the synthesis of metal oxide nanoparticle catalysts substantially impacted product analysis during photocatalytic studies. We confirmed these experimental results via hybrid Density Functional Theory computations of the materials' electronic properties to evaluate their viability as photocatalysts for nitrogen reduction. This nitrogen ligand contamination, and subsequent interference in photocatalytic studies, is avoidable through the careful design of synthetic pathways that exclude nitrogen-containing constituents. This result highlights the urgent need for careful evaluation of catalyst synthesis protocols, as contamination by nitrogen-containing ligands may go unnoticed since the presence of nitrogen is often not detected or probed., Comment: 13 pages, 3 figures

Let $p$ be prime and $e,k$ be positive integers, and let $R = {\mathbb Z}_p[\sqrt[e]{p}]$. We calculate the Waring numbers $g_R(k)$ for many values of $p, e$, and $k$, and investigate how the Waring numbers for $p=2$ change as $e$ and $k$ vary., Comment: 18 pages, 3 figures

Enterprise chatbots, powered by generative AI, are emerging as key applications to enhance employee productivity. Retrieval Augmented Generation (RAG), Large Language Models (LLMs), and orchestration frameworks like Langchain and Llamaindex are crucial for building these chatbots. However, creating effective enterprise chatbots is challenging and requires meticulous RAG pipeline engineering. This includes fine-tuning embeddings and LLMs, extracting documents from vector databases, rephrasing queries, reranking results, designing prompts, honoring document access controls, providing concise responses, including references, safeguarding personal information, and building orchestration agents. We present a framework for building RAG-based chatbots based on our experience with three NVIDIA chatbots: for IT/HR benefits, financial earnings, and general content. Our contributions are three-fold: introducing the FACTS framework (Freshness, Architectures, Cost, Testing, Security), presenting fifteen RAG pipeline control points, and providing empirical results on accuracy-latency tradeoffs between large and small LLMs. To the best of our knowledge, this is the first paper of its kind that provides a holistic view of the factors as well as solutions for building secure enterprise-grade chatbots.", Comment: 8 pages, 6 figures, 2 tables, Preprint submission to ACM CIKM 2024

We present the first detailed radio study of the galaxy cluster XMMXCS J2215.9-1738 at $z$ = 1.46 using MeerKAT $L$-band (1.3 GHz) observations. We combine our radio observation with archival optical and infrared data to investigate the star formation and AGN population within $R_{200}$ ($R = $ 0.8 Mpc) of the cluster centre. Using three selection criteria; the radio luminosity, the far-infrared radio ratio ($q_{\rm{IR}}$) and the mid-infrared colour, we distinguish galaxies with radio emission predominantly powered by star formation from that powered by AGNs. We selected 24 cluster members within $R_{\rm{200}}$ in the MeerKAT image based on either their photometric or spectroscopic redshift. We classified 12/24 ($50\%$) as galaxies whose radio emission is dominated by star-formation activity, 6/24 ($25\%$) as intermediate star-forming galaxies and 6/24 ($25\%$) as AGN-dominated galaxies. Using the radio continuum luminosities of the star-forming cluster galaxies, we estimated an integrated star formation rate (SFR) value of 1700 $\pm$ 330 M$_{\odot}$yr$^{-1}$ within $R_{200}$. We derived a mass-normalized integrated SFR value of $(570 \pm 110) \times 10^{-14}$ yr$^{-1}$. This supports previous observational and theoretical studies that indicated a rapid increase in star formation activity within the core of high-redshift clusters. We also show that the high AGN fraction within the cluster core is consistent with previous cluster observations at $z >$ 1.5.

Robots often use feature-based image tracking to identify their position in their surrounding environment; however, feature-based image tracking is prone to errors in low-textured and poorly lit environments. Specifically, we investigate a scenario where robots are tasked with exploring the surface of the Moon and are required to have an accurate estimate of their position to be able to correctly geotag scientific measurements. To reduce localization error, we complement traditional feature-based image tracking with ultra-wideband (UWB) distance measurements between the robots. The robots use an advanced mesh-ranging protocol that allows them to continuously share distance measurements amongst each other rather than relying on the common "anchor" and "tag" UWB architecture. We develop a decentralized multi-robot coordination algorithm that actively plans paths based on measurement line-of-sight vectors amongst all robots to minimize collective localization error. We then demonstrate the emergent behavior of the proposed multi-robot coordination algorithm both in simulation and hardware to lower a geometry-based uncertainty metric and reduce localization error., Comment: This work has been submitted to the IEEE for possible publication

Deep learning approaches for animal re-identification have had a major impact on conservation, significantly reducing the time required for many downstream tasks, such as well-being monitoring. We propose a method called Recurrence over Video Frames (RoVF), which uses a recurrent head based on the Perceiver architecture to iteratively construct an embedding from a video clip. RoVF is trained using triplet loss based on the co-occurrence of individuals in the video frames, where the individual IDs are unavailable. We tested this method and various models based on the DINOv2 transformer architecture on a dataset of meerkats collected at the Wellington Zoo. Our method achieves a top-1 re-identification accuracy of $49\%$, which is higher than that of the best DINOv2 model ($42\%$). We found that the model can match observations of individuals where humans cannot, and our model (RoVF) performs better than the comparisons with minimal fine-tuning. In future work, we plan to improve these models by using pre-text tasks, apply them to animal behaviour classification, and perform a hyperparameter search to optimise the models further., Comment: Presented as a poster at the CV4Animals Workshop, CVPR 2024

Optically addressable spin-based quantum sensors enable nanoscale measurements of temperature, magnetic field, pH, and other physical properties of a system. Advancing the sensors beyond proof-of-principle demonstrations in living cells and multicellular organisms towards reliable, damage-free quantum sensing poses three distinct technical challenges. First, spin-based quantum sensing requires optical accessibility and microwave delivery. Second, any microelectronics must be biocompatible and designed for imaging living specimens. Third, efficient microwave delivery and temperature control are essential to reduce unwanted heating and to maintain an optimal biological environment. Here, we present the Quantum Biosensing Chip (Q-BiC), which facilitates microfluidic-compatible microwave delivery and includes on-chip temperature control. We demonstrate the use of Q-BiC in conjunction with nanodiamonds containing nitrogen vacancy centers to perform optically detected magnetic resonance in living systems. We quantify the biocompatibility of microwave excitation required for optically detected magnetic resonance both in vitro in HeLa cells and in vivo in the nematode Caenorhabditis elegans for temperature measurements and determine the microwave-exposure range allowed before detrimental effects are observed. In addition, we show that nanoscale quantum thermometry can be performed in immobilised but non-anaesthetised adult nematodes with minimal stress. These results enable the use of spin-based quantum sensors without damaging the biological system under study, facilitating the investigation of the local thermodynamic and viscoelastic properties of intracellular processes.

We investigate modifications to Bayesian Optimization for a resource-constrained setting of sequential experimental design where changes to certain design variables of the search space incur a switching cost. This models the scenario where there is a trade-off between evaluating more while maintaining the same setup, or switching and restricting the number of possible evaluations due to the incurred cost. We adapt two process-constrained batch algorithms to this sequential problem formulation, and propose two new methods: one cost-aware and one cost-ignorant. We validate and compare the algorithms using a set of 7 scalable test functions in different dimensionalities and switching-cost settings for 30 total configurations. Our proposed cost-aware hyperparameter-free algorithm yields comparable results to tuned process-constrained algorithms in all settings we considered, suggesting some degree of robustness to varying landscape features and cost trade-offs. This method starts to outperform the other algorithms with increasing switching-cost. Our work broadens out from other recent Bayesian Optimization studies in resource-constrained settings that consider a batch setting only. While the contributions of this work are relevant to the general class of resource-constrained problems, they are particularly relevant to problems where adaptability to varying resource availability is of high importance

The tight relationship between infrared luminosity (L$_\mathrm{TIR}$) and 1.4 GHz radio continuum luminosity (L$_\mathrm{1.4GHz}$) has proven useful for understanding star formation free from dust obscuration. Infrared emission in star-forming galaxies typically arises from recently formed, dust-enshrouded stars, whereas radio synchrotron emission is expected from subsequent supernovae. By leveraging the wealth of ancillary far-ultraviolet - far-infrared photometry from the Deep Extragalactic VIsible Legacy Survey (DEVILS) and Galaxy and Mass Assembly (GAMA) surveys, combined with 1.4 GHz observations from the MeerKAT International GHz Tiered Extragalactic Exploration (MIGHTEE) survey and Deep Investigation of Neutral Gas Origins (DINGO) projects, we investigate the impact of timescale differences between far-ultraviolet - far-infrared and radio-derived star formation rate (SFR) tracers. We examine how the SED-derived star formation histories (SFH) of galaxies can be used to explain discrepancies in these SFR tracers, which are sensitive to different timescales. Galaxies exhibiting an increasing SFH have systematically higher L$_\mathrm{TIR}$ and SED-derived SFRs than predicted from their 1.4 GHz radio luminosity. This indicates that insufficient time has passed for subsequent supernovae-driven radio emission to accumulate. We show that backtracking the SFR(t) of galaxies along their SED-derived SFHs to a time several hundred megayears prior to their observed epoch will both linearise the SFR-L$_\mathrm{1.4GHz}$ relation and reduce the overall scatter. The minimum scatter in the SFR(t)-L$_\mathrm{1.4GHz}$ is reached at 200 - 300 Myr prior, consistent with theoretical predictions for the timescales required to disperse the cosmic ray electrons responsible for the synchrotron emission., Comment: 23 pages, 13 figures, 2 tables. Accepted for publication in MNRAS

The transport of molecules through biological and synthetic nanopores is governed by multiple stochastic processes that lead to noisy, fluctuating currents. Disentangling the characteristics of different noise-generating mechanisms is central to better understanding molecular transport at a fundamental level but is extremely challenging in molecular systems, due to their complexity and relative experimental inaccessibility. Here, we construct a colloidal model microfluidic system for the experimental measurement of particle currents, where the governing physical properties are directly controllable and particle dynamics directly observable, unlike in the molecular case. Currents of hard spheres fluctuate due to the random arrival times of particles into the channel and the distribution of particle speeds within the channel, which results in characteristic scalings in the power spectral density. We rationalise these scalings by quantitatively comparing to a model for shot noise with a finite transit time, extended to include the distribution of particle speeds. Particle velocity distributions sensitively reflect the confining geometry and we interpret and model these in terms of the underlying fluid flow profiles. Finally, we explore the extent to which details of these distributions govern the form of the resulting power spectral density, thereby establishing concrete links between the power spectral density and underlying mechanisms for this experimental system. This paves the way for establishing a more systematic understanding of the links between characteristics of transport fluctuations and underlying molecular mechanisms in driven systems such as nanopores., Comment: The following article has been accepted by the Journal of Chemical Physics. Version 2 has been extended to include additional modelling and analysis

Numerous methods have been proposed for global navigation satellite system (GNSS) receivers to detect faulty GNSS signals. One such fault detection and exclusion (FDE) method is based on the mathematical concept of Euclidean distance matrices (EDMs). This paper outlines a greedy approach that uses an improved Euclidean distance matrix-based fault detection and exclusion algorithm. The novel greedy EDM FDE method implements a new fault detection test statistic and fault exclusion strategy that drastically simplifies the complexity of the algorithm over previous work. To validate the novel greedy EDM FDE algorithm, we created a simulated dataset using receiver locations from around the globe. The simulated dataset allows us to verify our results on 2,601 different satellite geometries. Additionally, we tested the greedy EDM FDE algorithm using a real-world dataset from seven different android phones. Across both the simulated and real-world datasets, the Python implementation of the greedy EDM FDE algorithm is shown to be computed an order of magnitude more rapidly than a comparable greedy residual FDE method while obtaining similar fault exclusion accuracy. We provide discussion on the comparative time complexities of greedy EDM FDE, greedy residual FDE, and solution separation. We also explain potential modifications to greedy residual FDE that can be added to alter performance characteristics., Comment: Submitted to NAVIGATION: Journal of the Institute of Navigation

This paper presents gnss_lib_py, a Python library used to parse, analyze, and visualize data from a variety of GNSS (Global Navigation Satellite Systems) data sources. The gnss_lib_py library's ease of use, modular capabilities, testing coverage, and extensive documentation make it an attractive tool not only for scientific and industry users wanting a quick, out-of-the-box solution but also for advanced GNSS users developing new GNSS algorithms. gnss_lib_py has already demonstrated its usefulness and impact through presentation in academic conferences, use in research papers, and adoption in graduate-level university course curricula., Comment: Submitted to the SoftwareX journal

In the current paradigm, high redshift radio halos are expected to be scarce due to inverse Compton energy losses and redshift dimming, which cause them to be intrinsically faint. This low occurrence fraction is predicted by cosmic ray electron turbulent re-acceleration models. To date, only a handful of radio halos have been detected at redshift z > 0.8. We report the MeerKAT detection of a radio halo hosted by a galaxy cluster ACT-CL J0329.2-2330 at z = 1.23, making it the highest redshift halo detected thus far. Using L-band and UHF-band observations, we derive a radio halo spectral index of $\alpha^{1.3GHz}_{0.8GHz}$ = 1.3 $\pm$ 0.4 and a radio power of P$_{1.4GHz}$ = (4.4 $\pm$ 1.5) $\times$ 10$^{24}$ W Hz$^{-1}$. This result further confirms that there is rapid magnetic field amplification in galaxy clusters at high redshift., Comment: 8 pages, 8 figures, and 2 tables

We consider whether conditions exist under which block-coordinate descent is asymptotically efficient in evolutionary multi-objective optimization, addressing an open problem. Block-coordinate descent, where an optimization problem is decomposed into $k$ blocks of decision variables and each of the blocks is optimized (with the others fixed) in a sequence, is a technique used in some large-scale optimization problems such as airline scheduling, however its use in multi-objective optimization is less studied. We propose a block-coordinate version of GSEMO and compare its running time to the standard GSEMO algorithm. Theoretical and empirical results on a bi-objective test function, a variant of LOTZ, serve to demonstrate the existence of cases where block-coordinate descent is faster. The result may yield wider insights into this class of algorithms., Comment: Accepted at GECCO 2024

Current clustering priors for deep latent variable models (DLVMs) require defining the number of clusters a-priori and are susceptible to poor initializations. Addressing these deficiencies could greatly benefit deep learning-based scRNA-seq analysis by performing integration and clustering simultaneously. We adapt the VampPrior (Tomczak & Welling, 2018) into a Dirichlet process Gaussian mixture model, resulting in the VampPrior Mixture Model (VMM), a novel prior for DLVMs. We propose an inference procedure that alternates between variational inference and Empirical Bayes to cleanly distinguish variational and prior parameters. Using the VMM in a Variational Autoencoder attains highly competitive clustering performance on benchmark datasets. Augmenting scVI (Lopez et al., 2018), a popular scRNA-seq integration method, with the VMM significantly improves its performance and automatically arranges cells into biologically meaningful clusters.

Better understanding the natural world is a crucial task with a wide range of applications. In environments with close proximity between humans and animals, such as zoos, it is essential to better understand the causes behind animal behaviour and what interventions are responsible for changes in their behaviours. This can help to predict unusual behaviours, mitigate detrimental effects and increase the well-being of animals. There has been work on modelling the dynamics behind swarms of birds and insects but the complex social behaviours of mammalian groups remain less explored. In this work, we propose a method to build behavioural models using causal structure discovery and graph neural networks for time series. We apply this method to a mob of meerkats in a zoo environment and study its ability to predict future actions and model the behaviour distribution at an individual-level and at a group level. We show that our method can match and outperform standard deep learning architectures and generate more realistic data, while using fewer parameters and providing increased interpretability., Comment: 9 pages, 7 figures, accepted as an extended abstract and poster at AAMAS 2024

Crystal plasticity models are a powerful tool for predicting the deformation behaviour of polycrystalline materials accounting for the underlying grain morphology and texture. These models typically have a large number of parameters, an understanding of which is required to effectively calibrate and apply the model. This study presents a structured framework for the global sensitivity analysis, using Sobol indices, of the effect of crystal plasticity parameters on model outputs. Due to the computational cost of evaluating the crystal plasticity model multiple times within a finite element framework, a Gaussian process regression surrogate model was constructed and used to conduct the sensitivity analysis. Influential parameters from the sensitivity analysis were carried forward for calibration using both a local Nelder-Mead and global differential evolution optimisation algorithm. The results show that the surrogate based global sensitivity analysis is able to efficiently identify influential crystal plasticity parameters and parameter combinations. With a reduced parameter set, both the Nelder-Mead algorithm with multiple runs and the differential evolution algorithm were able to find an optimal parameter set which gives a close calibration of a simulated tensile curve to experimental data. However, only the differential evolutionary algorithm was able to reliably find the global optimum due to the presence of local minima in the calibration objective function.