Accurately deciphering spatial domains for spatially resolved transcriptomics with stCluster.

Spatial transcriptomics provides valuable insights into gene expression within the native tissue context, effectively merging molecular data with spatial information to uncover intricate cellular relationships and tissue organizations. In this context, deciphering cellular spatial domains becomes essential for revealing complex cellular dynamics and tissue structures. However, current methods encounter challenges in seamlessly integrating gene expression data with spatial information, resulting in less informative representations of spots and suboptimal accuracy in spatial domain identification. We introduce stCluster, a novel method that integrates graph contrastive learning with multi-task learning to refine informative representations for spatial transcriptomic data, consequently improving spatial domain identification. stCluster first leverages graph contrastive learning technology to obtain discriminative representations capable of recognizing spatially coherent patterns. Through jointly optimizing multiple tasks, stCluster further fine-tunes the representations to be able to capture complex relationships between gene expression and spatial organization. Benchmarked against six state-of-the-art methods, the experimental results reveal its proficiency in accurately identifying complex spatial domains across various datasets and platforms, spanning tissue, organ, and embryo levels. Moreover, stCluster can effectively denoise the spatial gene expression patterns and enhance the spatial trajectory inference. The source code of stCluster is freely available at https://github.com/hannshu/stCluster.

Collaborative deep learning improves disease-related circRNA prediction based on multi-source functional information.

Emerging studies have shown that circular RNAs (circRNAs) are involved in a variety of biological processes and play a key role in disease diagnosing, treating and inferring. Although many methods, including traditional machine learning and deep learning, have been developed to predict associations between circRNAs and diseases, the biological function of circRNAs has not been fully exploited. Some methods have explored disease-related circRNAs based on different views, but how to efficiently use the multi-view data about circRNA is still not well studied. Therefore, we propose a computational model to predict potential circRNA-disease associations based on collaborative learning with circRNA multi-view functional annotations. First, we extract circRNA multi-view functional annotations and build circRNA association networks, respectively, to enable effective network fusion. Then, a collaborative deep learning framework for multi-view information is designed to get circRNA multi-source information features, which can make full use of the internal relationship among circRNA multi-view information. We build a network consisting of circRNAs and diseases by their functional similarity and extract the consistency description information of circRNAs and diseases. Last, we predict potential associations between circRNAs and diseases based on graph auto encoder. Our computational model has better performance in predicting candidate disease-related circRNAs than the existing ones. Furthermore, it shows the high practicability of the method that we use several common diseases as case studies to find some unknown circRNAs related to them. The experiments show that CLCDA can efficiently predict disease-related circRNAs and are helpful for the diagnosis and treatment of human disease.

scMultiGAN: cell-specific imputation for single-cell transcriptomes with multiple deep generative adversarial networks.

The emergence of single-cell RNA sequencing (scRNA-seq) technology has revolutionized the identification of cell types and the study of cellular states at a single-cell level. Despite its significant potential, scRNA-seq data analysis is plagued by the issue of missing values. Many existing imputation methods rely on simplistic data distribution assumptions while ignoring the intrinsic gene expression distribution specific to cells. This work presents a novel deep-learning model, named scMultiGAN, for scRNA-seq imputation, which utilizes multiple collaborative generative adversarial networks (GAN). Unlike traditional GAN-based imputation methods that generate missing values based on random noises, scMultiGAN employs a two-stage training process and utilizes multiple GANs to achieve cell-specific imputation. Experimental results show the efficacy of scMultiGAN in imputation accuracy, cell clustering, differential gene expression analysis and trajectory analysis, significantly outperforming existing state-of-the-art techniques. Additionally, scMultiGAN is scalable to large scRNA-seq datasets and consistently performs well across sequencing platforms. The scMultiGAN code is freely available at https://github.com/Galaxy8172/scMultiGAN.

A network-based method for brain disease gene prediction by integrating brain connectome and molecular network.

Brain disease gene identification is critical for revealing the biological mechanism and developing drugs for brain diseases. To enhance the identification of brain disease genes, similarity-based computational methods, especially network-based methods, have been adopted for narrowing down the searching space. However, these network-based methods only use molecular networks, ignoring brain connectome data, which have been widely used in many brain-related studies. In our study, we propose a novel framework, named brainMI, for integrating brain connectome data and molecular-based gene association networks to predict brain disease genes. For the consistent representation of molecular-based network data and brain connectome data, brainMI first constructs a novel gene network, called brain functional connectivity (BFC)-based gene network, based on resting-state functional magnetic resonance imaging data and brain region-specific gene expression data. Then, a multiple network integration method is proposed to learn low-dimensional features of genes by integrating the BFC-based gene network and existing protein-protein interaction networks. Finally, these features are utilized to predict brain disease genes based on a support vector machine-based model. We evaluate brainMI on four brain diseases, including Alzheimer's disease, Parkinson's disease, major depressive disorder and autism. brainMI achieves of 0.761, 0.729, 0.728 and 0.744 using the BFC-based gene network alone and enhances the molecular network-based performance by 6.3% on average. In addition, the results show that brainMI achieves higher performance in predicting brain disease genes compared to the existing three state-of-the-art methods.

Enhancing discoveries of molecular QTL studies with small sample size using summary statistic imputation.

Quantitative trait locus (QTL) analyses of multiomic molecular traits, such as gene transcription (eQTL), DNA methylation (mQTL) and histone modification (haQTL), have been widely used to infer the functional effects of genome variants. However, the QTL discovery is largely restricted by the limited study sample size, which demands higher threshold of minor allele frequency and then causes heavy missing molecular trait-variant associations. This happens prominently in single-cell level molecular QTL studies because of sample availability and cost. It is urgent to propose a method to solve this problem in order to enhance discoveries of current molecular QTL studies with small sample size. In this study, we presented an efficient computational framework called xQTLImp to impute missing molecular QTL associations. In the local-region imputation, xQTLImp uses multivariate Gaussian model to impute the missing associations by leveraging known association statistics of variants and the linkage disequilibrium (LD) around. In the genome-wide imputation, novel procedures are implemented to improve efficiency, including dynamically constructing a reused LD buffer, adopting multiple heuristic strategies and parallel computing. Experiments on various multiomic bulk and single-cell sequencing-based QTL datasets have demonstrated high imputation accuracy and novel QTL discovery ability of xQTLImp. Finally, a C++ software package is freely available at https://github.com/stormlovetao/QTLIMP.

DFinder: a novel end-to-end graph embedding-based method to identify drug-food interactions.

MOTIVATION: Drug-food interactions (DFIs) occur when some constituents of food affect the bioaccessibility or efficacy of the drug by involving in drug pharmacodynamic and/or pharmacokinetic processes. Many computational methods have achieved remarkable results in link prediction tasks between biological entities, which show the potential of computational methods in discovering novel DFIs. However, there are few computational approaches that pay attention to DFI identification. This is mainly due to the lack of DFI data. In addition, food is generally made up of a variety of chemical substances. The complexity of food makes it difficult to generate accurate feature representations for food. Therefore, it is urgent to develop effective computational approaches for learning the food feature representation and predicting DFIs. RESULTS: In this article, we first collect DFI data from DrugBank and PubMed, respectively, to construct two datasets, named DrugBank-DFI and PubMed-DFI. Based on these two datasets, two DFI networks are constructed. Then, we propose a novel end-to-end graph embedding-based method named DFinder to identify DFIs. DFinder combines node attribute features and topological structure features to learn the representations of drugs and food constituents. In topology space, we adopt a simplified graph convolution network-based method to learn the topological structure features. In feature space, we use a deep neural network to extract attribute features from the original node attributes. The evaluation results indicate that DFinder performs better than other baseline methods. AVAILABILITY AND IMPLEMENTATION: The source code is available at https://github.com/23AIBox/23AIBox-DFinder. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Design of an image-based BRDF measurement method using a catadioptric multispectral capture and a real-time Lambert calibration.

By utilizing a catadioptric system and a calibration Lambertian sample, a compact measurement method of bidirectional reflectance distribution function (BRDF) has been proposed for rapid and accurate measurement. With the help of an ellipsoidal dome mirror, a hyperboloid mirror, and a high-resolution camera, spatial reflectance distributions from reflected directions with a large field of view (FOV) can be obtained. The built-in Lambertian standard allows for real-time calibration to account for fluctuations in the illumination spectrum, effectively reducing the measurement drift and achieving a high accuracy. Moreover, a multispectral camera captures images at 8 spectral bands for accurate spectral color reconstruction from different directions. To verify the method, a prototype capable of fast, high-resolution measurements with a large FOV has been developed for characterizing the scattering properties of objects. It achieves a measured angular range up to 160°. Multispectral BRDF data for each sample can be obtained within 5 minutes with an angular resolution of less than 0.6°. Eight ceramic samples with different colors were selected for the verification of measurement accuracy, and their mean relative bias of BRDF measurement was found to be as low as 2.5%.

Mapping-based design method for high-quality integral projection system.

A general method for designing an integral projection system is proposed, including optical design and digital preprocessing based on the mapping within the projection system. The per-pixel mapping between the sub-images and the integral projection image is generated by incorporating an integral projection imaging model as well as the ray data of all sub-channels. By tracing rays for sparsely sampled field points of the central sub-channel and constructing the mapping between the central sub-channel and other sub-channels, the efficient acquisition of ray data for all sub-channels is achieved. The sub-image preprocessing pipeline is presented to effectively address issues such as overlapping misalignment, optical aberrations, inhomogeneous illumination, and their collective contribution. An integral projection optical system with a field of view (FOV) of 80°, an F-number of 2, and uniform image performance is given as a design example. The ray tracing simulation results and quantitative analysis demonstrate that the proposed system yields distortion-free, uniformly illuminated, and high-quality integral projection images.

Ultracompact polarization multiplexing meta-combiner for augmented reality display.

Augmented reality (AR) display, as a next-generation innovative technology, is revolutionizing the ways of perceiving and communicating by overlaying virtual images onto real-world scenes. However, the current AR devices are often bulky and cumbersome, posing challenges for long-term wearability. Metasurfaces have flexible capabilities of manipulating light waves at subwavelength scales, making them as ideal candidates for replacing traditional optical elements in AR display devices. In this work, we propose and fabricate what we believe is a novel reflective polarization multiplexing gradient metasurface based on propagation phase principle to replace the optical combiner element in traditional AR display devices. Our designed metasurface exhibits different polarization modulations for reflected and transmitted light, enabling efficient deflection of reflected light while minimizing the impact on transmitted light. This work reveals the significant potential of metasurfaces in next-generation optical display systems and provides a reliable theoretical foundation for future integrated waveguide schemes, driving the development of next-generation optical display products towards lightweight and comfortable.

Polarized emission of Cs3Cu2I5 nanowires embedded in nanopores of an anodic aluminum oxide template.

Due to the intrinsic polarized emission property, polarized emissive materials with anisotropic nanostructures are expected to be potential substitutes for polarizers. Herein, by the template-assisted strategy, well-aligned lead-free metal halide Cs3Cu2I5 nanowire (NW) arrays are fabricated by evaporating the precursor ink in the anodic aluminum oxide (AAO) for polarized emission. The Cs3Cu2I5/AAO composite film emits highly polarized light with a degree of polarization (DOP) of 0.50. Furthermore, by changing the molar ratio of CsI/CuI, the stability of Cs3Cu2I5 precursor inks is improved. Finally, an ultraviolet (UV) light-emitting diode (LED) is adopted to pump the composite film to achieve a blue LED device. The reported Cs3Cu2I5/AAO composite film with highly polarized light emissions will have great potential for polarized emission applications such as liquid crystal display backlights, waveguides, and lasers.

Design of compact off-axis freeform imaging systems based on optical-digital joint optimization.

Using a freeform optical surface can effectively reduce the imaging system weight and volume while maintaining good performance and advanced system specifications. But it is still very difficult for traditional freeform surface design when ultra-small system volume or ultra-few elements are required. Considering the images generated by the system can be recovered by digital image processing, in this paper, we proposed a design method of compact and simplified off-axis freeform imaging systems using optical-digital joint design process, which fully integrates the design of a geometric freeform system and the image recovery neural network. This design method works for off-axis nonsymmetric system structure and multiple freeform surfaces with complicated surface expression. The overall design framework, ray tracing, image simulation and recovery, and loss function establishment are demonstrated. We use two design examples to show the feasibility and effect of the framework. One is a freeform three-mirror system with a much smaller volume than a traditional freeform three-mirror reference design. The other is a freeform two-mirror system whose element number is reduced compared with the three-mirror system. Ultra-compact and/or simplified freeform system structure as well as good output recovered images can be realized.

Effect of surrounding objects in the adapting scene on chromatic adaptation.

Most of the existing chromatic adaptation transforms (CATs) were developed for flat uniform stimuli presented in a uniform background, which substantially simplifies the complexity of the real scene by excluding surrounding objects from the viewing field. The impact of the background complexity, in terms of the spatial properties of the objects surrounding the stimulus, on chromatic adaptation is ignored in most CATs. This study systematically investigated how the background complexity and color distribution affect the adaptation state. Achromatic matching experiments were conducted in an immersive lighting booth, with the illumination varying in chromaticity and the adapting scene varying in surrounding objects. Results show that compared to the uniform adapting field, increasing the scene complexity can significantly improve the degree of adaptation for the Planckian illuminations with low CCT levels. In addition, the achromatic matching points are substantially biased by the color of the surrounding object, implying the interactive effect of the illumination color and the dominant scene color on the adapting white point.

DCPNet: a dual-channel parallel deep neural network for high quality computer-generated holography.

Recent studies have demonstrated that a learning-based computer-generated hologram (CGH) has great potential for real-time, high-quality holographic displays. However, most existing algorithms treat the complex-valued wave field as a two-channel spatial domain image to facilitate mapping onto real-valued kernels, which does not fully consider the computational characteristics of complex amplitude. To address this issue, we proposed a dual-channel parallel neural network (DCPNet) for generating phase-only holograms (POHs), taking inspiration from the double phase amplitude encoding method. Instead of encoding the complex-valued wave field in the SLM plane as a two-channel image, we encode it into two real-valued phase elements. Then the two learned sub-POHs are sampled by the complementary 2D binary grating to synthesize the desired POH. Simulation and optical experiments are carried out to verify the feasibility and effectiveness of the proposed method. The simulation results indicate that the DCPNet is capable of generating high-fidelity 2k POHs in 36 ms. The optical experiments reveal that the DCPNet has excellent ability to preserve finer details, suppress speckle noise and improve uniformity in the reconstructed images.

Automatic control method for freeform surface shapes.

Designing freeform optics with high degrees of freedom can improve their optical performances; however, there are high requirements for controlling the surface shapes of such optics. Optical designers need to add constraints to the optimization process and make repeated adjustments to ensure the manufacturability of these shapes; this process is cumbersome and relies heavily on the experience of the designer. In this study, an automatic control method for freeform surface shapes is proposed. By adding an outer loop to the optimization process, the principal curvature and sag departure of the sampling points are gradually controlled during the optimization cycle based on the system requirements and surface evaluation results. The method was implemented in CODE V and successfully applied to a design example in freeform prism optics.

Complex amplitude modulated holographic display system based on polarization grating.

We propose a holographic display system for complex amplitude modulation (CAM) using a phase-only spatial light modulator (SLM) and two polarization gratings (PG). The two sub-holograms of the complex-amplitude computed generated hologram (CGH) are loaded in different regions of SLM. Two diffractive components couple in space after longitudinal migration from the double PGs, and finally interfered through the line polarizer. The influence of the system error on the reconstructed image quality is analyzed, which provides a theoretical assessment for adding pre-compensation to CGH to compensate the system error. Moreover, on the base of the proposed system, a large depth of field and enlarged display area display is realized and the real-time display can be achieved because of the analytical complex-amplitude computed generated hologram. The optical experimental results show that the proposed system has high energy efficiency, and can provide high-quality holographic display with a large depth of field and enlarged display area.

Fabrication of an ultra-high quality MgF2 micro-resonator for a single soliton comb generation.

Crystalline micro-resonators are attractive for a wide range of applications due to their extremely high quality (Q) factor. In this paper, we develop a semi-automatic method for fabricating ultra-high Q-factor MgF2 crystalline micro-resonators. By utilizing a force feedback sensor and corresponding control, we made a semi-automatic precision grind-and-polishing machine, and successfully fabricated trapezoid MgF2 resonators with diameter of 9.5 mm and a root mean square surface roughness of 0.26 nm. The maximum difference of peaks and valleys is about 1.5 nm. The Q-factor was characterized to be 9.24 × 109at 1550 nm by the cavity ring-down spectroscopy. A single soliton optical frequency comb was generated by pumping the microcavity with 150 mW optical power.

Holographic display using layered computer-generated volume hologram.

The spatial frequency of the reconstructed image of planar computer-generated hologram(CGH) is limited by the sampling interval and the lack of thickness. To break through this limitation of planar CGH, we propose a new computer-generated volume hologram(CGVH) for full-color dynamic holographic three-dimensional(3D) display, and an iteration-free layered CGVH generation method. The proposed CGVH is equivalent to a volume hologram sampled discretely in three directions. The generation method employs the layered angular spectral diffraction to calculate the light field in the layered CGVH, and then encodes it into a CGVH. Numerical simulation results show that the CGVH can accurately reconstruct full-color 3D objects, where better imaging quality, more concentrated diffraction energy, denser reconstructed spatial frequency information, and larger viewing angle are achieved. The proposed CGVH is expected to be applied to realize dynamic modulation, wavelength multiplexing, and angle multiplexing in various optical fields in the future.

Design and fabrication method of holographic waveguide near-eye display with 2D eye box expansion.

Augmented reality near-eye display (AR-NED) technology has attracted enormous interests for its widespread potential applications. In this paper, two-dimensional (2D) holographic waveguide integrated simulation design and analysis, holographic optical elements (HOEs) exposure fabrication, prototype performance evaluation and imaging analysis are completed. In the system design, a 2D holographic waveguide AR-NED integrated with a miniature projection optical system is presented to achieve a larger 2D eye box expansion (EBE). A design method for controlling the luminance uniformity of 2D-EPE holographic waveguide by dividing the two thicknesses of HOEs is proposed, which is easy to fabricate. The optical principle and design method of the HOE-based 2D-EBE holographic waveguide are described in detail. In the system fabrication, laser exposure fabrication method of eliminating stray light for HOEs is proposed, and a prototype system is fabricated and demonstrated. The properties of the fabricated HOEs and the prototype are analyzed in detail. The experimental results verified that the 2D-EBE holographic waveguide has a diagonal field of view (FOV) of 45°, an ultra-thin thickness of 1 mm, and an eye box of 16 mm × 13 mm at an eye relief (ERF) of 18 mm, the MTF values of different FOVs at different 2D-EPE positions can be better than 0.2 at 20 lp/mm, and the whole luminance uniformity is 58%.

Convolutional symmetric compressed look-up-table method for 360° dynamic color 3D holographic display.

In this paper, we propose a convolutional symmetric compressed look-up-table (CSC-LUT) method to accelerate computer-generated hologram (CGH) computation based on the Fresnel diffraction theory and LUT. The proposed method can achieve one-time high-quality fast generation of color holograms by utilizing dynamic convolution operation, which is divided three processes. Firstly, the pre-calculated data of maximum horizontal modulation factor is compressed in 1D array by coordinate symmetry. Then, the test object is resampled to satisfy convolutional translation invariance. Finally, the dynamic convolution operation is used to simplify CGH computation process rather than the point-by-point computation. Numerical simulation and optical experimental results show that our proposed method can achieve faster computation speed, higher reconstruction quality and wider application compared to conventional SC-LUT method. The further optimization method for parallel acceleration on the GPU framework can achieve real-time (>24fps) color holographic display corresponding to three perspectives of a 3D scene.

Depth of field expansion method for integral imaging based on diffractive optical element and CNN.

In lens-based display systems, lens aberrations and depth of field (DoF) limitation often lead to blurring and distortion of reconstructed images; Meanwhile, expanding the display DoF will face a trade-off between horizontal resolution and axial resolution, restricting the achievement of high-resolution and large DoF three-dimensional (3D) displays. To overcome these constraints and enhance the DoF and resolution of reconstructed scenes, we propose a DoF expansion method based on diffractive optical element (DOE) optimization and image pre-correction through a convolutional neural network (CNN). This method applies DOE instead of the conventional lens and optimizes DOE phase distribution using the Adam algorithm, achieving depth-invariant and concentrated point spread function (PSF) distribution throughout the entire DoF range; Simultaneously, we utilize a CNN to pre-correct the original images and compensate for the image quality reduction introduced by the DOE. The proposed method is applied to a practical integral imaging system, we effectively extend the DoF of the DOE to 400 mm, leading to a high-resolution 3D display in multiple depth planes. To validate the effectiveness and practicality of the proposed method, we conduct numerical simulations and optical experiments.