Aleksandra Badura

Aleksandra Badura

Erasmus University MC, Rotterdam, The Netherlands

Aleksandra Badura


Cerebellar pathologies are often found in people with autism, and cerebellar lesions at birth lead to a significant autism-risk. These observations brought me to investigate the role of the cerebellum in autism. I tested the hypothesis that early-life cerebellar insult leads to lasting deficits in adulthood and demonstrated that such early-life manipulations lead to a variety of long-lasting social and flexible behavior deficits in adult mice (Badura et al. 2018). In 2018, I was awarded a VIDI starting grant from The Netherlands Organization for Health Research and Development (ZonMw) to work on understanding the cerebello-cerebral networks underlying shared autistic traits. In my laboratory, we develop novel automated methods to test behavioural phenotypes of autism-mouse models (Wahl et al. 2022). We have also shown that TSC1 haploinsufficiency, which is often comorbid with autism, leads to the aberrant development of inhibitory cells in the developing cerebellum (Serra et al. 2022). In the last three years, my laboratory has developed extensive collaborations with Departments of Immunology to study the causal role of primary immunodeficiency on increased autism risk (Kaiser et al. 2022). By combining mouse models with the patients’ data my laboratory aims to explain how mutations can leads to both autism and immune deficits. In 2023 together with my collaborators we were awarded an NWA-ORC 2022 grant of 5.3 mln Euro to study how genetic, neurophysiological and behavioral sex differences contribute to the presence and severity of autism characteristics, and to distinguish these biological factors from potential diagnostic bias. The long-term goal of the SCANNER project is to improve the autism diagnostic process by introducing novel sex-sensitive solutions applicable in clinical settings and eHealth care.

Description of the general focus of the symposium "Automated analysis of behavior"

Nowadays, an increasing number of automated systems to monitor behaviour of laboratory mice in specially designed arenas, as well as in the homecage environment are being developed. Although, most of these methods integrate technologies grounded in video analysis, they are often limited by arbitrarily defined parameters and highly-cost softwares. Concurrently, these technologies usually depend on the experimentator's presence and subjective perception, which most likely impact interpretation of the obtained results. While commercially available programmes still remains prevalent in preclinical research, a compelling prospect is noticed in Artificial Intelligence (AI)-based tools. Such systems provide a strong support in automatization of measuring complex behavioral patterns in mice, enable human input minimalization and may detect subtle, but often crucial changes in rodents phenotypes.

In my scientific panel, I aim to concentrate on the automatization and objectivity within behavioral studies related to mice. I would like to focus on predominant Artificial Intelligence (AI)-based tools, to highlight differences in their interfaces and user accessibility. My goal is to raise a discussion about the diversity of AI-driven tools outputs, in relevance to the research aim.


Current phenotyping approaches for murine autism models often focus on one selected behavioral feature, making the translation onto a spectrum of autistic characteristics in humans challenging. Furthermore, sex and environmental factors are rarely considered. I will discuss our latest work in which we aimed to capture the full spectrum of behavioral manifestations in three autism mouse models to develop a “behavioral fingerprint” that takes environmental and sex influences under consideration. To this end, we employed a wide range of classical standardized behavioral tests and two multi-parametric behavioral assays: the Live Mouse Tracker and Motion Sequencing (MoSeq), on male and female Shank2, Tsc1 and Pcp2-Tsc1 mutant mice raised in standard or enriched environments. We found that most behavioral phenotypes were dependent on sex- and environment. Furthermore, multi-parametric behavioral assays enabled far more accurate classification of experimental groups compared to classical tests. Together, our results provide a complete phenotypic description of all tested groups, suggesting multi-parametic assays can capturing the entire spectrum of the heterogenous phenotype in autism mouse models.

Our partners
The Neuronus Neuroscience Forum website uses cookies in accordance with the Privacy Policy. We ask for your consent to use anonymous data to improve your experience of our website. Privacy Policy