RBRasneer Sonia Bains

Rasneer Sonia Bains

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Rasneer Sonia Bains has a total of 166 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Using Machine Learning to Automate the Analysis of an Olfactory Habituation‐Dishabituation Task in Mice

    by S. Boyanova, M. H. Correa, R. S. Bains, F. K. Wiseman

    Introduction: Improving the efficiency and accuracy of annotation and extraction of performance data from mouse behavioral tasks will improve both the throughput and scientific value of preclinical research. Methods: Here, we present and validate an automated pipeline for the annotation and quantification of performance in a mouse olfactory habituation‐dishabituation task, using a single side‐view camera, resulting in occluded body parts. We created a pipeline for task analysis, combining DeepLabCut (DLC) for pose‐estimation and SimBA, for behavioral classification to automatically quantify odor interaction (sniffing time) in a three‐odor (water, familiar mouse social odor, novel mouse social odor) variant of the task. We used a subset of previously published, fully manually annotated datasets to train the models and unseen videos from the same study to validate the utility of our machine learning (ML) pipeline. Results and Conclusion: Our analysis pipeline estimated behavioral performance in the task with high accuracy, and the data produces similar technical and biological results to manual methods when analyzed by linear mixed modeling. Thus, we validated the utility of our customized pipeline for the automated scoring of this mouse sensory task, and we discuss its strengths and limitations.

  • Open Access

    Critical uncertainties in preclinical research: Navigating trust, technology, and ethics

    by Indrek Heinla, Vootele Voikar, Hilary Gates, Rasneer Sonia Bains, Sara Wells, Silvia Mandillo, Maša Čater, Kristi Kukk, Lars Lewejohann, Dragan Hrnčić, et al.

  • Open Access

    Using machine learning to automate the analysis of an olfactory habituation-dishabituation task in mice

    by S. Boyanova, M. H. Correa, R. S. Bains, F. K. Wiseman

    Introduction: Improving the efficiency and accuracy of annotation and extraction of performance data from mouse behavioural tasks will improve both the throughput and scientific value of preclinical research. Methods: Here, we present and validate an automated pipeline for the annotation and quantification of performance in a mouse olfactory habituation-dishabituation task, using a single side-view camera, resulting in occluded body parts. We created a pipeline for task analysis, combining DeepLabCut, for pose-estimation, and SimBA, for behavioural classification to automatically quantify odour interaction (sniffing time) in a three-odour (water, familiar mouse social odour, novel mouse social odour) variant of the task. We used a subset of previously published, fully manually annotated datasets to train the models and unseen videos from the same study to validate the utility of our machine learning pipeline. Results and conclusion: Our analysis pipeline estimated behavioural performance in the task with high accuracy, and the data produces similar technical and biological results to manual methods when analysed by linear mixed modelling. Thus, we validated the utility of our new pipeline for the automated scoring of this mouse sensory task.

  • Open Access

    Hierarchical Deep Learning for Abnormality Classification in Mouse Skeleton Using Multiview X-Ray Images: Convolutional Autoencoders Versus ConvNeXt

    by Muhammad M. Jawaid, Rasneer S. Bains, Sara Wells, James M. Brown

    Single-view-based anomaly detection approaches present challenges due to the lack of context, particularly for multi-label problems. In this work, we demonstrate the efficacy of using multiview image data for improved classification using a hierarchical learning approach. Using 170,958 images from the International Mouse Phenotyping Consortium (IMPC) repository, a specimen-wise multiview dataset comprising 54,046 specimens was curated. Next, two hierarchical classification frameworks were developed by customizing ConvNeXT and a convolutional autoencoder (CAE) as CNN backbones, respectively. The customized architectures were trained at three hierarchy levels with increasing anatomical granularity, enabling specialized layers to learn progressively more detailed features. At the top level (L1), multiview (MV) classification performed about the same as single views, with a high mean AUC of 0.95. However, using MV images in the hierarchical model greatly improved classification at levels 2 and 3. The model showed consistently higher average AUC scores with MV compared to single views such as dorsoventral or lateral. For example, at Level 2 (L2), the model divided abnormal cases into three subclasses, achieving AUCs of 0.65 for DV, 0.76 for LV, and 0.87 for MV. Then, at Level 3 (L3), it further divided these into ten specific abnormalities, with AUCs of 0.54 for DV, 0.59 for LV, and 0.82 for MV. A similar performance was achieved by the CAE-driven architecture, with mean AUCs of 0.87, 0.88, and 0.89 at Level 2 (L2) and 0.74, 0.78, and 0.81 at Level 3 (L3), respectively, for DV, LV, and MV views. The overall results demonstrate the advantage of multiview image data coupled with hierarchical learning for skeletal abnormality detection in a multi-label context.


  • Open Access

    Multi-modal comparative phenotyping of knock-in mouse models of frontotemporal dementia/amyotrophic lateral sclerosis

    by Sevda Boyanova, Gareth Banks, Tatiana V. Lipina, Rasneer Sonia Bains, Hamish Forrest, Michelle Stewart, Mireia Carcolé, Carmelo Milioto, Adrian M. Isaacs, Sara E. Wells, et al.

    Amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) are progressive adult-onset neurodegenerative diseases with overlapping pathological and genetic origins. They are caused by multiple underlying mechanisms leading to a common collection of clinical features that occur in a spectrum. Here, we report side-by-side longitudinal behavioural, cognitive and sensory phenotyping of two mouse models of ALS/FTD, to determine which aspects of the disease they recapitulate. We used knock-in models, in which the endogenous mouse orthologues of the C9orf72 and TARDBP (encoding TDP-43) genes have been altered to model specific molecular aspects of ALS/FTD. We found that the C9orf72GR400/+ model exhibits age-related deficit in short-term memory and that parental genotype affects exploration activity in offspring. In the TardbpQ331K/Q331K model, we found age-related changes in weight, fat mass, locomotion and marble burying. In both models, we found no evidence of deficits in vision or olfactory habituation-dishabituation. These data provide new insight into genotype-phenotype relationships in these ALS/FTD mice, which can be used to inform model choice and experimental design in future research studies.


  • Open Access

    Too big to lose - a FAIR repository for biomedical data derived from home-cage monitoring

    by Rasneer Sonia Bains, Damien Huzard, James Edgar McCutcheon, Pawel Boguszewski, Davor Virag, Leonardo Restivo, Lars Lewejohann, Michael Ashby, Jan Rozman, Rivalan Marion, et al.

    Home-Cage Monitoring (HCM) systems have revolutionised the collection of continuous behavioural data in rodent research and offer unprecedented insights into animal behaviour, yet the full potential of these data remains largely untapped. This paper, developed within the European network COST Action TEATIME, argues for a collaborative approach to harness the power of HCM data through interdisciplinary partnerships with machine learning experts and data scientists.We identify challenges that persist in fully capturing and analysing the complete behavioural repertoire of rodents over extended periods and propose the development of FAIR-compliant (Findable, Accessible, Interoperable, and Reusable) repositories for HCM (meta)data.Standardised data formats and comprehensive metadata enable confident predictions and discoveries across diverse scientific domains. By facilitating data sharing and collaborative analysis, we aim to unlock novel insights into health and disease, ultimately enhancing research efficiency and reducing animal use.Establishing this infrastructure will not only advance our understanding of rodent behaviour but also pave the way for innovative applications of machine learning and big data analytics in biomedical research.


  • Open Access

    Enhanced mGluR1 function causes motor deficits and region-specific Purkinje cell dysfunction

    by Mohamed F. Ibrahim, Sevda Boyanova, Yin Chun Cheng, Clemence Ligneul, Rasneer S. Bains, Jason P. Lerch, Ed Mann, Peter L. Oliver, Esther B. E. Becker

    Spinocerebellar ataxias (SCAs) are autosomal dominantly inherited neurodegenerative disorders with no effective treatment. Aberrant signalling through the metabotropic glutamate receptor (mGluR1) has been implicated in several SCAs. However, whether disease is caused through decreased or increased mGluR1 signalling remains controversial. Here, we generate the first mouse model of enhanced mGluR1 function by introducing a gain-of-function mutation (p.Y792C) that causes SCA44 in the metabotropic glutamate receptor 1 (Grm1 ) gene. Grm1 mutant mice recapitulate key pathophysiological aspects of SCA, including progressive motor deficits, altered climbing fibre innervation and perturbed Purkinje cell spontaneous activity. We report that changes in synaptic innervation and intrinsic Purkinje cell activity upon overactive mGluR1 signalling manifest in a lobule- and disease-stage-specific manner. Our findings demonstrate that enhanced mGluR1 function is a direct and specific driver of Purkinje cell dysfunction and pathology and provide a mechanism for understanding the selective vulnerability of different Purkinje cell populations in SCA.

  • Open Access

    Long-read transcriptomic identification of synaptic adaptation to amyloid pathology in the App NL-G-F knock-in mouse model of the earliest phase of Alzheimer’s disease

    by Umran Yaman, Gareth Banks, Emil K. Gustavsson, Paige Mumford, Naciye Magusali, Orjona Stella Taso, Hannah Macpherson, Susana Carmona, Malgorzata Murray, Rasneer Sonia Bains, et al.

    Genome-wide association studies (GWAS) have identified a transcriptional network of Alzheimer’s disease (AD) risk genes that are primarily expressed in microglia and are associated with AD pathology. However, traditional short-read sequencers have limited our ability to fully characterize how GWAS variants exert their effects on gene expression regulation or alternative splicing in response to the pathology, particularly resulting in inaccurate detection of splicing. To address this gap, we utilized long-read RNA-sequencing (RNA-seq) in the App NL-G-F knock-in mouse model to identify changes in splicing and novel transcript isoforms in response to amyloid-β. We show that long-read RNA-seq can recapitulate the expected induction of microglial expressed risk genes such as Trem2 in response to amyloid-β at 9 months of age associated with ageing-dependent deficiencies in spatial short-term memory in the App NL-G-F knock-in mice. Our results not only identified novel splicing events and transcript isoform abundance in genes associated with AD, but also revealed the complex regulation of gene expression through splicing in response to amyloid plaques. Surprisingly, the regulation of alternative splicing in response to amyloid was seen in genes previously not identified as AD risk genes, expressed in microglia, neurons and oligodendrocytes, and included genes such as Syngr1 that modulate synaptic physiology. We saw alternative splicing in genes such as Ctsa , Clta , Dennd2a , Irf9 and Smad4 in mice in response to amyloid, and the orthologues of these genes also showed transcript usage changes in human AD brains. Our data suggests a model whereby induction of AD risk gene expression associated with microglial proliferation and activation is concomitant with alternative splicing in a different class of genes expressed by microglia and neurons, which act to adapt or preserve synaptic activity in response to amyloid-β during early stages of the disease. Our study provides new insights into the mechanisms and effects of the regulation of genes associated with amyloid pathology, which may ultimately enable better disease diagnosis, and improved tracking of disease progression. Additionally, our findings identify new therapeutic avenues for treatment of AD.


  • Hierarchical Multi-label Learning for Musculoskeletal Phenotyping in Mice

    by Muhammad Moazzam Jawaid, Sonia Rasneer Bains, Sara Wells, James M. Brown

  • Open Access

    Of Mice and Mates: Automated Classification and Modelling of Mouse Behaviour in Groups Using a Single Model Across Cages

    by Michael P. J. Camilleri, Rasneer S. Bains, Christopher K. I. Williams

    Behavioural experiments often happen in specialised arenas, but this may confound the analysis. To address this issue, we provide tools to study mice in the home-cage environment, equipping biologists with the possibility to capture the temporal aspect of the individual’s behaviour and model the interaction and interdependence between cage-mates with minimal human intervention. Our main contribution is the novel Global Behaviour Model (GBM) which summarises the joint behaviour of groups of mice across cages, using a permutation matrix to match the mouse identities in each cage to the model. In support of the above, we also (a) developed the Activity Labelling Module (ALM) to automatically classify mouse behaviour from video, and (b) released two datasets, ABODe for training behaviour classifiers and IMADGE for modelling behaviour.