Penna, B. | bioRxiv Science

bioRxiv2025

Harnessing Interpretable Deep Learning to Predict Meropenem Resistance in Klebsiella pneumoniae

Antimicrobial resistance constitutes an escalating global health threat, complicating therapeutic management and increasing morbidity and mortality. Deep learning approaches have emerged as promising tools for bacterial profiling based on omics data, particularly for predicting antimicrobial susceptibility from genomic information. This task relies on identifying genomic signatures associated with resistance mechanisms. Here, DeepMDC is introduced as a deep learning architecture designed for bacterial profiling using whole-genome data. Given that precise annotation at the gene or mutation level is often costly and ambiguous, phenotypic classification is formulated as a Multiple Instance Learning (MIL) problem, in which each genome is represented as a bag of instances with a single associated label. The core of DeepMDC is a Modern Hopfield Network that processes all open reading frames (ORFs), including small ones, derived from genomic data. A key feature of the architecture is its interpretability, enabled by attention mechanisms that facilitate biological insight and hypothesis generation. The model was evaluated against Klebsiella pneumoniae and four clinically relevant antibiotics (meropenem, cefepime, ceftazidime, and gentamicin), achieving strong performance in several metrics. Notably, genes associated with resistance consistently received high attention scores during inference, which validates the architecture and eventually may generate new hypotheses.

systems biology↗

bioRxiv2019

Dynamic encoding of social threat and spatial context in the hypothalamus

Territorial animals must be able to express social aggression or avoidance in a manner appropriate to spatial context and dominance status. Recent studies indicate that the ventromedial hypothalamus controls both innate aggression and avoidance, suggesting that it may encode an internal state of threat common to both behaviors. Here we used single unit in vivo calcium microendoscopy to identify neurons in the mouse ventromedial hypothalamus encoding social threat. Threat neurons were activated during social defeat as well as when the animal performed risk assessment. Unexpectedly, threat neurons were also activate in the chamber where the animal had been previously defeated and a distinct set of neurons emerged that were active in its home chamber, demonstrating the dynamic encoding of spatial context in the hypothalamus. Ensemble analysis of neural activity showed that social defeat induced a change in the encoding of social information and optogenetic activation of ventromedial hypothalamus neurons was able to elicit avoidance after, but not before social defeat, demonstrating a functional reorganization of the pathway by social experience. These findings reveal how instinctive behavior circuits in the hypothalamus dynamically encode spatial and sensory cues to drive adaptive social behaviors.

neuroscience↗