Current food crisis predictions are developed by the Famine Early Warning System Network, but they fail to classify the majority of food crisis outbreaks with model metrics of recall (0.23), precision (0.42), and f1 (0.30). In this work, using a World Bank dataset, classical and neural network (NN) machine learning algorithms were developed to predict food crises in 21 countries. The best classical logistic regression algorithm achieved a high level of significance (p < 0.001) and precision (0.75) but was deficient in recall (0.20) and f1 (0.32). Of particular interest, the classical algorithm indicated that the vegetation index and the food price index were both positively correlated with food crises. A novel method for performing an iterative multidimensional hyperparameter search is presented, which resulted in significantly improved performance when applied to this dataset. Four iterations were conducted, which resulted in excellent 0.96 for metrics of precision, recall, and f1. Due to this strong performance, the food crisis year was removed from the dataset to prevent immediate extrapolation when used on future data, and the modeling process was repeated. The best “no year” model metrics remained strong, achieving ≥0.92 for recall, precision, and f1 while meeting a 10% f1 overfitting threshold on the test (0.84) and holdout (0.83) datasets. The year-agnostic neural network model represents a novel approach to classify food crises and outperforms current food crisis prediction efforts.

Alterations in synaptic homeostasis are linked to cognitive and behavioural impairments in brain disorders. However, synaptic dysfunction in childhood dementia is poorly understood. Here, we generate human cortical circuits from induced pluripotent stem cells (iPSCs) derived from donors with Mucopolysaccharidosis Type IIIA (MPS IIIA), also known as Sanfilippo syndrome, a common form of childhood-onset dementia. Action potential firing capacity and morphology of MPS IIIA patient neurons in culture are similar to those of neurons from neurotypical donors. However, long-term neural maturation reveals excitation/inhibition imbalances caused by hyperactive excitatory synapses, disrupted network dynamics, and dysregulated gene expression linked to synaptic homeostasis. This study validates in vitro human neural models to detect neurophysiological phenotypes in childhood dementias and supports drug discovery strategies that target synaptic dysfunction to improve cognition in MPS IIIA and related brain disorders.

Childhood dementias are a group of paediatric neurodegenerative disorders characterised by neurocognitive decline, and in many cases underpinned by pathophysiological mechanisms similar to adult-onset dementias. In this study, we use patient-derived induced pluripotent stem cells (iPSCs) from children with one of the most prevalent childhood dementias, Mucopolysaccharidosis Type IIIA (MPS IIIA), also known as Sanfilippo syndrome. The derived cortical cultures exhibit lysosomal dysfunction, heparan sulfate accumulation, progressive neurodegeneration and astrocytic reactivity, recapitulating prototypical in-vivo phenotypes. Using a multimodal drug screening platform that integrates machine learning, high-content confocal imaging, single-nuclei transcriptomics and electrophysiology, we identify at least nine repurposed compounds that significantly mitigate these adverse effects within two weeks of treatment in vitro, demonstrating potential for rapid clinical translation. This human preclinical model for MPS IIIA, coupled with a robust multimodal therapeutic interrogation platform, serves as an exemplar for advancing drug discovery for childhood dementias and the broader neurodegenerative disease spectrum.