Spherical geometry, adaptive optics, and highly dense network of neurons bridging the eye with the visual cortex, are the primary features of human eyes which enable wide field-of-view (FoV), low aberration, excellent adaptivity, and preprocessing of perceived visual information. Therefore, fabricating spherical artificial eyes has garnered enormous scientific interest. However, fusing color vision, in-device preprocessing and optical adaptivity into spherical artificial eyes has always been a tremendous challenge. Herein, we demonstrate a bionic eye comprising tunable liquid crystal optics, and a hemispherical neuromorphic retina with filter-free color vision, enabled by wavelength dependent bidirectional synaptic photo-response in a metal-oxide nanotube/perovskite nanowire hybrid structure. Moreover, by tuning the color selectivity with bias, the device can reconstruct full color images. This work demonstrates a unique approach to address the color vision and optical adaptivity issues associated with artificial eyes that can bring them to a new level approaching their biological counterparts.

To cope with seasonal changes in the environment, organisms adapt their physiology and behavior. Although color perception varies among seasons, the underlying molecular basis and its physiological significance remain unclear. Here we show that dynamic plasticity in phototransduction regulates seasonal changes in color perception in medaka fish. Medaka are active and exhibit clear phototaxis in conditions simulating summer, but remain at the bottom of the tank and fail to exhibit phototaxis in conditions simulating winter. Mate preference tests using virtual fish created with computer graphics demonstrate that medaka are more attracted to orange-red-colored model fish in summer than in winter. Transcriptome analysis of the eye reveals dynamic seasonal changes in the expression of genes encoding photopigments and their downstream pathways. Behavioral analysis of photopigment-null fish shows significant differences from wild type, suggesting that plasticity in color perception is crucial for the emergence of seasonally regulated behaviors.Animal coloration and behavior can change seasonally, but it is unclear if visual sensitivity to color shifts as well. Here, Shimmura et al. show that medaka undergo seasonal behavioral change accompanied by altered expression of opsin genes, resulting in reduced visual sensitivity to mates during winter-like conditions.

Mammalian retinal metabolism favors aerobic glycolysis. However, the role of glycolytic metabolism in retinal morphogenesis remains unknown. We report that aerobic glycolysis is necessary for the early stages of retinal development. Taking advantage of an unbiased approach that combines the use of eye organoids and single-cell RNA sequencing, we identify specific glucose transporters and glycolytic genes in retinal progenitors. Next, we determine that the optic vesicle territory of mouse embryos displays elevated levels of glycolytic activity. At the functional level, we show that removal of Glucose transporter 1 and Lactate dehydrogenase A gene activity from developing retinal progenitors arrests eye morphogenesis. Surprisingly, we uncover that lactate-mediated upregulation of key eye-field transcription factors is controlled by the epigenetic modification of histone H3 acetylation through histone deacetylase activity. Our results identify an unexpected bioenergetic independent role of lactate as a signaling molecule necessary for mammalian eye morphogenesis.

The coding privilege of end-spectral hues (red and blue) in the early visual cortex has been reported in primates. However, the origin of such bias remains unclear. Here, we provide a complete picture of the end-spectral bias in visual system by measuring fMRI signals and spiking activities in macaques. The correlated end-spectral biases between the LGN and V1 suggest a subcortical source for asymmetric coding. Along the ventral pathway from V1 to V4, red bias against green peaked in V1 and then declined, whereas blue bias against yellow showed an increasing trend. The feedforward and recurrent modifications of end-spectral bias were further revealed by dynamic causal modeling analysis. Moreover, we found that the strongest end-spectral bias in V1 was in layer 4C[Formula: see text]. Our results suggest that end-spectral bias already exists in the LGN and is transmitted to V1 mainly through the parvocellular pathway, then embellished by cortical processing.

The mechanisms by which DNA alleles contribute to disease risk, drug response, and other human phenotypes are highly context-specific, varying across cell types and different conditions. Human induced pluripotent stem cells are uniquely suited to study these context-dependent effects but cell lines from hundreds or thousands of individuals are required. Village cultures, where multiple induced pluripotent stem lines are cultured and differentiated in a single dish, provide an elegant solution for scaling induced pluripotent stem experiments to the necessary sample sizes required for population-scale studies. Here, we show the utility of village models, demonstrating how cells can be assigned to an induced pluripotent stem line using single-cell sequencing and illustrating that the genetic, epigenetic or induced pluripotent stem line-specific effects explain a large percentage of gene expression variation for many genes. We demonstrate that village methods can effectively detect induced pluripotent stem line-specific effects, including sensitive dynamics of cell states.

Transgenesis of most insects currently relies on fluorescence markers. Here we establish a transformation marker system causing phenotypes visible to the naked eye due to changes in the color of melanin pigments, which are widespread in animals. Ubiquitous overexpression of arylalkylamine-N-acetyl transferase in the silkworm, Bombyx mori, changes the color of newly hatched first-instar larvae from black to a distinctive light brown color, and can be used as a molecular marker by directly connecting to baculovirus immediate early 1 gene promoter. Suppression of black pigmentation by Bm-arylalkylamine-N-acetyl transferase can be observed throughout the larval stages and in adult animals. Alternatively, overexpression in another gene, B. mori β-alanyl-dopamine synthetase (Bm-ebony), changes the larval body color of older instars, although first-instar larvae had normal dark coloration. We further show that ectopic Bm-arylalkylamine-N-acetyl transferase expression lightens coloration in ladybird beetle Harmonia axyridis and fruit fly Drosophila melanogaster, highlighting the potential usefulness of this marker for transgenesis in diverse insect taxa.

Major depressive disorder ranks as a major burden of disease worldwide, yet the current antidepressant medications are limited by frequent non-responsiveness and significant side effects. The lateral septum (LS) is thought to control of depression, however, the cellular and circuit substrates are largely unknown. Here, we identified a subpopulation of LS GABAergic adenosine A2A receptors (A2AR)-positive neurons mediating depressive symptoms via direct projects to the lateral habenula (LHb) and the dorsomedial hypothalamus (DMH). Activation of A2AR in the LS augmented the spiking frequency of A2AR-positive neurons leading to a decreased activation of surrounding neurons and the bi-directional manipulation of LS-A2AR activity demonstrated that LS-A2ARs are necessary and sufficient to trigger depressive phenotypes. Thus, the optogenetic modulation (stimulation or inhibition) of LS-A2AR-positive neuronal activity or LS-A2AR-positive neurons projection terminals to the LHb or DMH, phenocopied depressive behaviors. Moreover, A2AR are upregulated in the LS in two male mouse models of repeated stress-induced depression. This identification that aberrantly increased A2AR signaling in the LS is a critical upstream regulator of repeated stress-induced depressive-like behaviors provides a neurophysiological and circuit-based justification of the antidepressant potential of A2AR antagonists, prompting their clinical translation.

Covalent organic frameworks have shown considerable application potential and exceptional properties in the construction of stimulus-responsive materials. Here, we designed a sweat-responsive covalent organic framework film for material-based fingerprint liveness detection. When exposed to human sweat, the COFTPDA-TFPy film can transform from yellow to red. The COFTPDA-TFPy film, when touched by living fingers, can produce the naked-eye-identified fingerprint pattern through the sweat-induced color change, while artificial fake fingerprints cannot. This technique, which we named material-based liveness detection, can thus intuitively discern living fingers from fake fingerprints with a 100% accuracy rate. Additionally, the distribution of sweat pores on human skin can also be collected and analyzed by shortening the contact time. By merely washing them with ethanol, all the samples can be utilized again. This work inventively accomplished material-based liveness detection and naked-eye-identified sweat pore analysis and highlighted their potential for use in clinical research and personal identification.

Dysregulation of the alternative pathway (AP) of the complement system is a significant contributor to age-related macular degeneration (AMD), a primary cause of irreversible vision loss worldwide. Here, we assess the contribution of the liver-produced complement factor H-related 4 protein (FHR-4) to AMD initiation and course of progression. We show that FHR-4 variation in plasma and at the primary location of AMD-associated pathology, the retinal pigment epithelium/Bruch's membrane/choroid interface, is entirely explained by three independent quantitative trait loci (QTL). Using two distinct cohorts composed of a combined 14,965 controls and 20,741 cases, we ascertain that independent QTLs for FHR-4 are distinct from variants causally associated with AMD, and that FHR-4 variation is not independently associated with disease. Additionally, FHR-4 does not appear to influence AMD progression course among patients with disease driven predominantly by AP dysregulation. Modulation of FHR-4 is therefore unlikely to be an effective therapeutic strategy for AMD.

Blood vessels in the central nervous system (CNS) develop unique features, but the contribution of CNS neurons to regulating those features is not fully understood. We report that inhibiting spontaneous cholinergic activity or reducing starburst amacrine cell numbers prevents invasion of endothelial cells into the deep layers of the retina and causes blood-retinal-barrier (BRB) dysfunction in mice. Vascular endothelial growth factor (VEGF), which drives angiogenesis, and Norrin, a Wnt ligand that induces BRB properties, are decreased after activity blockade. Exogenous VEGF restores vessel growth but not BRB function, whereas stabilizing beta-catenin in endothelial cells rescues BRB dysfunction but not vessel formation. We further identify that inhibiting cholinergic activity reduces angiogenesis during oxygen-induced retinopathy. Our findings demonstrate that neural activity lies upstream of VEGF and Norrin, coordinating angiogenesis and BRB formation. Neural activity originating from specific neural circuits may be a general mechanism for driving regional angiogenesis and barrier formation across CNS development.

The ventromedial prefrontal-cortex (vmPFC) is known to contain expected value signals that inform our choices. But expected values even for the same stimulus can differ by task. In this study, we asked how the brain flexibly switches between such value representations in a task-dependent manner. Thirty-five participants alternated between tasks in which either stimulus color or motion predicted rewards. We show that multivariate vmPFC signals contain a rich representation that includes the current task state or context (motion/color), the associated expected value, and crucially, the irrelevant value of the alternative context. We also find that irrelevant value representations in vmPFC compete with relevant value signals, interact with task-state representations and relate to behavioral signs of value competition. Our results shed light on vmPFC's role in decision making, bridging between its role in mapping observations onto the task states of a mental map, and computing expected values for multiple states.

Mobile and wearable devices are increasingly reliant on near-infrared (NIR) covert illumination for facial recognition, eye-tracking or motion and depth sensing functions. However, these small devices offer limited spatial real estate that is typically already occupied by their full-area electronic color displays. Here, we report a transparent perovskite light-emitting diode (LED) that could be overlaid across a color display to provide an efficient and high-intensity NIR illumination. Our transparent devices are constructed with an ITO/AZO/PEIE/FAPbI3/poly-TPD/MoO3/Al/ITO/Ag/ITO architecture, and offer a high average transmittance of more than 55% across the visible spectral region. In particular, our Al/ITO/Ag/ITO top transparent electrode was designed to offer a combination low sheet resistance and low plasma damage upon electrode deposition. The devices emit at 799 nm with a high total external quantum efficiency of 5.7% at a current density of 5.3 mA cm-2 and a high radiance of 1.5 W sr-1 m-2, and possess a large functional device area of 120 mm2. The efficient performance is ideal for battery-powered wearable devices, and could enable advanced security and sensing features on future smart-watches, phones, gaming consoles and augmented or virtual reality headsets.

Near-eye display technology is a rapidly growing field owing to the recent emergence of augmented and mixed reality. Ultrafast response time, high resolution, high luminance, and a dynamic range for outdoor use are all important for non-pixelated, pupil-forming optics. The current mainstream technologies using liquid crystals and organic materials cannot satisfy all these conditions. Thus, finely patterned light-emissive solid-state devices with integrated circuits are often proposed to meet these requirements. In this study, we integrated several advanced technologies to design a prototype microscale light-emitting diode (LED) arrays using quantum dot (QD)-based color conversion. Wafer-scale epilayer transfer and the bond-before-pattern technique were used to directly integrate 5-µm-scale GaN LED arrays on a foreign silicon substrate. Notably, the lithography-level alignment with the bottom wafer opens up the possibility for ultrafast operation with circuit integration. Spectrally pure color conversion and solvent-free QD patterning were also achieved using an elastomeric topographical mask. Self-assembled monolayers were applied to selectively alter the surface wettability for a completely dry process. The final emissive-type LED array integrating QD, GaN, and silicon technology resulted in a 1270 PPI resolution that is far beyond the retinal limit.

Recently, metasurfaces composed of artificially fabricated subwavelength structures have shown remarkable potential for the manipulation of light with unprecedented functionality. Here, we first demonstrate a metasurface application to realize a compact near-eye display system for augmented reality with a wide field of view. A key component is a see-through metalens with an anisotropic response, a high numerical aperture with a large aperture, and broadband characteristics. By virtue of these high-performance features, the metalens can overcome the existing bottleneck imposed by the narrow field of view and bulkiness of current systems, which hinders their usability and further development. Experimental demonstrations with a nanoimprinted large-area see-through metalens are reported, showing full-color imaging with a wide field of view and feasibility of mass production. This work on novel metasurface applications shows great potential for the development of optical display systems for future consumer electronics and computer vision applications.

Rapid, visual detection of pathogen nucleic acids has broad applications in infection management. Here we present a modular detection platform, termed enzyme-assisted nanocomplexes for visual identification of nucleic acids (enVision). The system consists of an integrated circuit of enzyme-DNA nanostructures, which function as independent recognition and signaling elements, for direct and versatile detection of pathogen nucleic acids from infected cells. The built-in enzymatic cascades produce a rapid color readout for the naked eye; the assay is thus fast (<2 h), sensitive (<10 amol), and readily quantified with smartphones. When implemented on a configurable microfluidic platform, the technology demonstrates superior programmability to perform versatile computations, for detecting diverse pathogen targets and their virus-host genome integration loci. We further design the enVision platform for molecular-typing of infections in patient endocervical samples. The technology not only improves the clinical inter-subtype differentiation, but also expands the intra-subtype coverage to identify previously undetectable infections.

Though consistently shown to detect mammographically occult cancers, breast ultrasound has been noted to have high false-positive rates. In this work, we present an AI system that achieves radiologist-level accuracy in identifying breast cancer in ultrasound images. Developed on 288,767 exams, consisting of 5,442,907 B-mode and Color Doppler images, the AI achieves an area under the receiver operating characteristic curve (AUROC) of 0.976 on a test set consisting of 44,755 exams. In a retrospective reader study, the AI achieves a higher AUROC than the average of ten board-certified breast radiologists (AUROC: 0.962 AI, 0.924 ± 0.02 radiologists). With the help of the AI, radiologists decrease their false positive rates by 37.3% and reduce requested biopsies by 27.8%, while maintaining the same level of sensitivity. This highlights the potential of AI in improving the accuracy, consistency, and efficiency of breast ultrasound diagnosis.

Diffusion magnetic resonance imaging (dMRI) is commonly used to assess the tissue and cellular substructure of the human brain. In the white matter, myelinated axons are the principal neural elements that shape dMRI through the restriction of water diffusion; however, in the gray matter the relative contributions of myelinated axons and other tissue features to dMRI are poorly understood. Here we investigate the determinants of diffusion in the cerebral cortex. Specifically, we ask whether myelinated axons significantly shape dMRI fractional anisotropy (dMRI-FA), a measure commonly used to characterize tissue properties in humans. We compared ultra-high resolution ex vivo dMRI data from the brain of a marmoset monkey with both myelin- and Nissl-stained histological sections obtained from the same brain after scanning. We found that the dMRI-FA did not match the spatial distribution of myelin in the gray matter. Instead dMRI-FA was more closely related to the anisotropy of stained tissue features, most prominently those revealed by Nissl staining and to a lesser extent those revealed by myelin staining. Our results suggest that unmyelinated neurites such as large caliber apical dendrites are the primary features shaping dMRI measures in the cerebral cortex.

It has been shown that human judgements of trustworthiness are based on subtle processing of specific facial features. However, it is not known if this ability is a specifically human function, or whether it is shared among primates. Here we report that macaque monkeys (Macaca Mulatta and Macaca Fascicularis), like humans, display a preferential attention to trustworthiness-associated facial cues in computer-generated human faces. Monkeys looked significantly longer at faces categorized a priori as trustworthy compared to untrustworthy. In addition, spatial sequential analysis of monkeys' initial saccades revealed an upward shift with attention moving to the eye region for trustworthy faces while no change was observed for the untrustworthy ones. Finally, we found significant correlations between facial width-to-height ratio- a morphometric feature that predicts trustworthiness' judgments in humans - and looking time in both species. These findings suggest the presence of common mechanisms among primates for first impression of trustworthiness.

An unresolved issue in contemporary biomedicine is the overwhelming number and diversity of complex images that require annotation, analysis and interpretation. Recent advances in Deep Learning have revolutionized the field of computer vision, creating algorithms that compete with human experts in image segmentation tasks. However, these frameworks require large human-annotated datasets for training and the resulting "black box" models are difficult to interpret. In this study, we introduce Kartezio, a modular Cartesian Genetic Programming-based computational strategy that generates fully transparent and easily interpretable image processing pipelines by iteratively assembling and parameterizing computer vision functions. The pipelines thus generated exhibit comparable precision to state-of-the-art Deep Learning approaches on instance segmentation tasks, while requiring drastically smaller training datasets. This Few-Shot Learning method confers tremendous flexibility, speed, and functionality to this approach. We then deploy Kartezio to solve a series of semantic and instance segmentation problems, and demonstrate its utility across diverse images ranging from multiplexed tissue histopathology images to high resolution microscopy images. While the flexibility, robustness and practical utility of Kartezio make this fully explicable evolutionary designer a potential game-changer in the field of biomedical image processing, Kartezio remains complementary and potentially auxiliary to mainstream Deep Learning approaches.

Reaching sub-millisecond 3D tracking of individual molecules in living cells would enable direct measurements of diffusion-limited macromolecular interactions under physiological conditions. Here, we present a 3D tracking principle that approaches the relevant regime. The method is based on the true excitation point spread function and cross-entropy minimization for position localization of moving fluorescent reporters. Tests on beads moving on a stage reaches 67 nm lateral and 109 nm axial precision with a time resolution of 0.84 ms at a photon count rate of 60 kHz; the measurements agree with the theoretical and simulated predictions. Our implementation also features a method for microsecond 3D PSF positioning and an estimator for diffusion analysis of tracking data. Finally, we successfully apply these methods to track the Trigger Factor protein in living bacterial cells. Overall, our results show that while it is possible to reach sub-millisecond live-cell single-molecule tracking, it is still hard to resolve state transitions based on diffusivity at this time scale.