We address the problem of inferring the number of independently blinking fluorescent light emitters, when only their combined intensity contributions can be observed. This problem occurs regularly in light microscopy of objects smaller than the diffraction limit, where one wishes to count the number of fluorescently labeled subunits. Our proposed solution directly models the photophysics of the system, as well as the blinking kinetics of the fluorescent emitters as a fully differentiable hidden Markov model, estimating a posterior distribution of the total number of emitters. We show that our model is more accurate and increases the range of countable subunits by a factor of 2 compared to current state-of-the-art methods. Furthermore, we demonstrate that our model can be used to investigate the effect of blinking kinetics on counting ability and therefore can inform optimal experimental conditions.

Many animals navigate using optic flow, detecting rotational image velocity differences between their eyes to adjust direction. Forward locomotion produces strong symmetric translational optic flow that can mask these differences, yet the brain efficiently extracts these binocular asymmetries for course control. In Drosophila melanogaster, monocular horizontal system neurons facilitate detection of binocular asymmetries and contribute to steering. To understand these functions, we reconstructed horizontal system cells' central network using electron microscopy datasets, revealing convergent visual inputs, a recurrent inhibitory middle layer and a divergent output layer projecting to the ventral nerve cord and deeper brain regions. Two-photon imaging, GABA receptor manipulations and modeling, showed that lateral disinhibition reduces the output's translational sensitivity while enhancing its rotational selectivity. Unilateral manipulations confirmed the role of interneurons and descending outputs in steering. These findings establish competitive disinhibition as a key circuit mechanism for detecting rotational motion during translation, supporting navigation in dynamic environments.

Preprint: https://doi.org/10.1101/2023.08.06.552150

Artificial neural networks learn faster if they are initialized well. Good initializations can generate high-dimensional macroscopic dynamics with long timescales. It is not known if biological neural networks have similar properties. Here we show that the eigenvalue spectrum and dynamical properties of large-scale neural recordings in mice (two-photon and electrophysiology) are similar to those produced by linear dynamics governed by a random symmetric matrix that is critically normalized. An exception was hippocampal area CA1: population activity in this area resembled an efficient, uncorrelated neural code, which may be optimized for information storage capacity. Global emergent activity modes persisted in simulations with sparse, clustered or spatial connectivity. We hypothesize that the spontaneous neural activity reflects a critical initialization of whole-brain neural circuits that is optimized for learning time-dependent tasks.

Summary Solute carrier family 11 member 1 (SLC11A1) is critical for host resistance to diverse intracellular pathogens. During infection, SLC11A1 limits Salmonella’s access to iron, zinc, and magnesium, but only magnesium deprivation significantly impairs Salmonella replication. To understand the unexpected minor impact of iron, we determined Salmonella’s iron access in infected SLC11A1-deficient and normal mice. Using reporter strains and mass spectrometry of Salmonella purified from the spleen, we found that SLC11A1 caused growth-restricting iron deprivation in a subset of Salmonella. Volume electron microscopy revealed that another Salmonella subset circumvented iron restriction by targeting iron-rich endosomes in macrophages degrading red blood cells (erythrophagocytosis). These iron-replete bacteria dominated overall Salmonella growth, masking the effects of the other Salmonella subset’s iron deprivation. Thus, SLC11A1 effectively sequesters iron, but heterogeneous Salmonella populations partially bypass this nutritional immunity by targeting iron-rich tissue microenvironments.

To survive, animals must be able quickly infer the state of their surroundings. For example, to successfully escape an approaching predator, prey must quickly estimate the direction of approach from incoming sensory stimuli and guide their behavior accordingly. Such rapid inferences are particularly challenging because the animal has only a brief window of time to gather sensory stimuli, and yet the accuracy of inference is critical for survival. Due to evolutionary pressures, nervous systems have likely evolved effective computational strategies that enable accurate inferences under strong time limitations. Traditionally, the relationship between the speed and accuracy of inference has been described by the “speed-accuracy tradeoff” (SAT), which quantifies how the average performance of an ideal observer improves as the observer has more time to collect incoming stimuli. While this trial-averaged description can reasonably account for individual inferences made over long timescales, it does not capture individual inferences on short timescales, when trial-to-trial variability gives rise to diverse patterns of error dynamics. We show that an ideal observer can exploit this single-trial structure by adaptively tracking the dynamics of its belief about the state of the environment, which enables it to speed its own inferences and more reliably track its own error, but also causes it to violate the SAT. We show that these features can be used to improve overall performance during rapid escape. The resulting behavior qualitatively reproduces features of escape behavior in the fruit fly Drosophila melanogaster, whose escapes have presumably been highly optimized by natural selection.

Circulating tumor cells (CTCs) are critical biomarkers for predicting therapy response and survival in breast cancer patients. Multicellular CTC clusters exhibit enhanced metastatic potential, yet their detection and characterization are constrained by low frequency in blood samples and reliance on labor-intensive manual analysis. Advancing these methods could significantly improve prognostic evaluation and therapeutic strategies.Leveraging FDA-approved CellSearch technology and single-cell sequencing, we analyzed 2, 853 blood specimens, longitudinally collected from 1358 patients with advanced cancer (breast, prostate, etc) and other diseases. Integrating machine learning and deep learning tools, we developed a novel CTCpose platform to automate detection and analysis of CTCs, immune cells, and their interactions. Using artificial intelligence (AI)-driven image analysis, we extracted over 270 cellular and nuclear features including intensity, morphometry, fourier shape, gradient/edge, and haralick of cytokeratin, CD45, and DAPI expression patterns, enabling precise characterization of CTCs, white blood cells (WBCs), CTC clusters, and their interactions with immune cells (WBCs).The CTCpose platform enabled automated identification of CTCs, WBCs, homotypic CTC clusters, heterogenous CTC-WBC clusters, and immune cell clusters, providing comprehensive insights into cell morphology, biomarker expression, and spatial organization. These features correlated with patient survival, disease progression, and treatment response. Our findings highlight the clinical significance of CTC-immune cell interactions and dynamic alterations of CTCs (singles and clusters) and underscore their potential in stratifying patients into distinct risk categories.This study demonstrates the transformative potential of deep learning in overcoming limitations of traditional CTC detection methods and integrating imaging data with large cohorts of patient data. By automating and enhancing the analysis of CTC-immune cell interactions, we present a robust framework for developing predictive models with direct clinical relevance. This work opens avenues for personalized treatment strategies, underscoring the impact of AI in advancing precision oncology.Yuanfei Sun, Joshua R. Squires, Andrew Hoffmann, Youbin Zhang, Allegra Minor, Anmol Singh, David Scholten, Chengsheng Mao, Yuan Luo, Deyu Fang, William J. Gradishar, Massimo Cristofanilli, Carsen Stringer, Huiping Liu. Deep learning enables automated detection of circulating tumor cell-immune cell interactions with prognostic insights in cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2420.

Comprehensive mapping of neural connections is essential for understanding brain function. Existing automated methods for connectome reconstruction from high-resolution images of brain tissue introduce errors that require extensive and time-consuming manual correction, a critical bottleneck in the field. To address this, we developed PATHFINDER, an AI system that segments volumetric image data, identifies potential ways to assemble neuron fragments, and evaluates the plausibility of resulting shapes to reconstruct complete neurons. Using a dataset of all axons in an IBEAM-mSEM volume of mouse cortex, we show that PATHFINDER reduces the error rate in axon reconstruction by an order of magnitude over previous state of the art, leading to an improvement in proofreading throughput of up to 84× relative to prior estimates in the context of a whole mouse brain. By drastically reducing the manual effort required for analysis, this advance unlocks the potential for both large-scale connectome mapping and routine investigation of smaller volumes.

During brain development, synapses are initially formed in excess and are later eliminated in an activity-dependent manner, with weak synapses being preferentially removed. Previous studies identified glia as mediators of synapse removal, but it is unclear how glia specifically target weak synapses. Here we show that, in the developing mouse visual pathway, inhibiting synaptic transmission induces postsynaptic activation of caspase-3. Caspase-3 is essential for synapse elimination driven by both spontaneous and experience-dependent neural activity. Synapse weakening-induced caspase-3 activation determines the specificity of synapse elimination mediated by microglia but not astrocytes. Furthermore, in a mouse model of Alzheimer’s disease, caspase-3 deficiency protects against synapse loss induced by amyloid-β deposition. Our results reveal caspase-3 activation as a key step in activity-dependent synapse elimination during development and synapse loss in neurodegeneration.

bioRxiv preprint: https://doi.org/10.1101/2024.08.02.606316

The use of fluorescent sensors for functional imaging has revolutionized the study of organellar Ca2+ signaling. However, understanding the dynamic interplay between intracellular Ca2+ sinks and sources requires bright, photostable and multiplexed measurements in each signaling compartment of interest to dissect the origins and destinations of Ca2+ fluxes. We introduce a new toolkit of chemigenetic indicators based on HaloCaMP, optimized to report Ca2+ dynamics in the endoplasmic reticulum (ER) and mitochondria of mammalian cells and neurons. Both ER-HaloCaMP and Mito-HaloCaMP present high brightness and responsiveness, and the use of different HaloTag ligands enables tunable red and far-red emission when quantifying organelle Ca2+ dynamics, expanding significantly multiplexing capacities of Ca2+ signaling. The improved brightness of ER-HaloCaMP using either red or far-red HaloTag ligands enabled measuring ER Ca2+ fluxes in axons of neurons, in which the ER is formed by a tiny tubule of 30-60 nanometers of diameter that impeded measurements with previous red ER Ca2+ sensors. When measuring ER Ca2+ fluxes in activated neuronal dendritic spines of cultured neurons, ER-HaloCaMP presented increased photostability compared to the gold-standard ER Ca2+ sensor in the field, ER-GCaMP6-210, while presenting the same responsiveness. On the other hand, Mito-HaloCaMP presented higher responsiveness than current red sensors, and enabled the first measurements of mitochondrial Ca2+ signaling in far-red in cell lines and primary neurons. As a proof-of-concept, we used 3-plex multiplexing to quantify interorganellar Ca2+ signaling. We show that effective transfer of Ca2+ from the ER to mitochondria depends on the ER releasing a critical amount of Ca2+. When this threshold is not met, the mobilized Ca2+ is diverted to the cytosol instead. Our new toolkit provides an expanded palette of bright, photostable and responsive organellar Ca2+ sensors, which will facilitate future studies of intracellular Ca2+ signaling.