Biology subjects

Kong, J.

Publications and source records attributed to Kong, J..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

mRNA lipid nanoparticle-incorporated nanofiber-hydrogel composite generates a local immunostimulatory niche for cancer immunotherapy

Hydrogel materials have emerged as versatile platforms for various biomedical applications. Notably, the engineered nanofiber-hydrogel composite (NHC) has proven effective in mimicking the soft tissue extracellular matrix, facilitating substantial recruitment of host immune cells and the formation of a local immunostimulatory microenvironment. Leveraging this feature, here we report an mRNA lipid nanoparticle (LNP)-incorporated NHC microgel matrix, termed LiNx, by incorporating LNPs loaded with mRNA encoding tumour antigens. Harnessing the potent transfection efficiency of LNPs in antigen-presenting cells (APCs), LiNx demonstrates remarkable immune cell recruitment, antigen expression and presentation, and cellular interaction. These attributes collectively create an immunostimulating milieu and yield a potent immune response achievable with a single dose, comparable to the conventional three-dose LNP immunization regimen. Further investigations reveal that the LiNx not only generates heightened Th1 and Th2 responses but also elicits a distinctive Type 17 T helper cell-mediated response pivotal for bolstering antitumour efficacy. Our findings elucidate the mechanism underlying LiNxs role in potentiating antigen-specific immune responses, presenting a new strategy for cancer immunotherapy.

bioengineering↗

bioRxiv2024

A Systematic Comparison of Single-Cell Perturbation Response Prediction Models

Predicting single-cell transcriptional responses to perturbations is central to dissecting gene regulation and accelerating therapeutic design, yet the field lacks a rigorous, task-spanning assessment of model behavior. We present a large-scale benchmark of 13 representative methods and baselines across 25 datasets spanning diverse perturbation modalities and species, including two primary immune-cell drug-response resources. We evaluated three core tasks--generalization to unseen single-gene perturbations, prediction of combinatorial interactions, and transfer across cell types--using 24 metrics covering expression-level accuracy, relative changes, differential expression recovery, and distributional similarity. Across tasks, performance depended strongly on perturbation effect size and evaluation perspective: expression-level agreement was highest for small-effect perturbations resembling controls, whereas delta- and DE-based metrics improved with larger effects, providing clearer signals. Models shared a conservative bias, with fine-tuned foundation models compressing variance and underestimating synergistic effects in combinations. PerturbNet showed superior recovery of DE signatures in Tasks 1 and 2, while no method consistently generalized across cell types in Task 3, where biological consistency dominated outcomes. This benchmark establishes current methodological limits, clarifies that different metrics probe distinct biological signals rather than redundant summaries of the same prediction problem, and provides a foundation for developing virtual-cell models that more faithfully capture heterogeneous perturbation responses. Teaser: A large-scale benchmark reveals strengths and limits of models predicting single-cell perturbation responses.

bioinformatics↗

bioRxiv2024

SegFinder: an automated tool for identifying RNA virus genome segments through co-occurrence in multiple sequenced samples

Metagenomic sequencing has expanded the RNA virosphere, but many identified viral genomes remain incomplete, especially for segmented viruses. Traditional methods relying on sequence homology struggle to identify highly divergent segments and group them confidently within a single virus species. To address this, we developed a new bioinformatic tool - SegFinder - that identifies virus genome segments based on their common co-occurrence at similar abundance within segmented viruses. SegFinder successfully re-discovered all segments from a test data set of individual mosquito transcriptomes, which was also used to establish parameter thresholds for reliable segment identification. Using these optimal parameters, we applied SegFinder to 858 libraries from eight metagenomic sequencing projects, including vertebrates, invertebrates, plants, and environmental samples. Furthermore, we identified 108 (excluding RdRP) unique viral genome segments, of which 55 were novel and 32 showed no recognizable sequence homology to known sequences but which were verified by the presence of conserved sequences at the genome termini. SegFinder is also able to identify segmented genome structures in viruses previously considered to be predominantly unsegmented, and in doing so expanded the number of known families and orders of segmented RNA viruses, making it a valuable tool in an era of large-scale parallel sequencing.

bioinformatics↗

bioRxiv2024

VirID: Beyond Virus Discovery - An Integrated Platform for Comprehensive RNA Virus Characterization

RNA viruses exhibit vast phylogenetic diversity and can significantly impact public health and agriculture. However, current bioinformatics tools for viral discovery from metagenomic data frequently generate false positive virus results, overestimate viral diversity, and misclassify virus sequences. Additionally, current tools often fail to determine virus-host associations, which hampers investigation of the potential threat posed by a newly detected virus. To address these issues we developed VirID, a software tool specifically designed for the discovery and characterization of RNA viruses from metagenomic data. The basis of VirID is a comprehensive RNA-dependent RNA polymerase (RdRP) database to enhance a workflow that includes RNA virus discovery, phylogenetic analysis, and phylogeny-based virus characterization. Benchmark tests on a simulated data set demonstrated that VirID had high accuracy in profiling viruses and estimating viral richness. In evaluations with real-world samples, VirID was able to identity RNA viruses of all type, but also provided accurate estimations of viral genetic diversity and virus classification, as well as comprehensive insights into virus associations with humans, animals, and plants. VirID therefore offers a robust tool for virus discovery and serves as a valuable resource in basic virological studies, pathogen surveillance, and early warning systems for infectious disease outbreaks.

bioinformatics↗

bioRxiv2024

CREEPING STEM1 Regulates Directional Auxin Transport for Lodging Resistance in Soybean

Soybean, a staple crop on a global scale, frequently encounters challenges due to lodging under high planting densities, which results in significant yield losses. Despite extensive research, the fundamental genetic mechanisms governing lodging resistance in soybeans remain elusive. In this study, we identify and characterize the Creeping Stem 1 (CS1) gene, which plays a crucial role in conferring lodging resistance in soybeans. The CS1 gene encodes a HEAT-repeat protein that modulates hypocotyl gravitropism by regulating amyloplast sedimentation. Functional analysis reveals that the loss of CS1 activity disrupts polar auxin transport, vascular bundle development, and the biosynthesis of cellulose and lignin, ultimately leading to premature lodging and aberrant root development. Conversely, increasing CS1 expression significantly enhances lodging resistance and improves yield under conditions of high planting density. Our findings shed light on the genetic mechanisms that underlie lodging resistance in soybeans and highlight the potential of CS1 as a valuable target for genetic engineering to improve crop lodging resistance and yield.

plant biology↗

bioRxiv2024

Augmented Reality and Cross-Device Interaction for Seamless Integration of Physical and Digital Scientific Papers

Researchers face the challenge of efficiently navigating vast scientific literature while valuing printed papers in the digital age. Printed materials facilitate deeper engagement and comprehension, leading to superior exam performance and enhanced retention. However, existing digital tools often need to pay more attention to the needs of researchers who value the tactile benefits of printed documents. In response to this gap, we introduce AR-PaperSync, a transformative solution that leverages Augmented Reality (AR) and cross-device interaction technology. AR-PaperSync seamlessly integrates the physical experience of printed papers with the interactive capabilities of digital tools. Researchers can effortlessly navigate inline citations, manage saved references, and synchronize reading notes across mobile, desktop, and printed paper formats. Our user-centric approach, informed by in-depth interviews with six researchers, ensures that AR-PaperSync is tailored to its target users needs. A comprehensive user study involving 28 participants evaluated AR-PaperSyncs significantly improved efficiency, accuracy, and cognitive load in academic reading tasks compared to conventional methods. These findings suggest that AR-PaperSync enhances the reading experience of printed scientific papers and provides a seamless integration of physical and digital reading environments for researchers.

scientific communication and education↗

bioRxiv2024

Integrated germline and somatic features reveal divergent immune pathways driving ICB response

Immune Checkpoint Blockade (ICB) has revolutionized cancer treatment, however mechanisms determining patient response remain poorly understood. Here we used machine learning to predict ICB response from germline and somatic biomarkers and interpreted the learned model to uncover putative mechanisms driving superior outcomes. Patients with higher T follicular helper infiltrates were robust to defects in the class-I Major Histocompatibility Complex (MHC-I). Further investigation uncovered different ICB responses in MHC-I versus MHC-II neoantigen reliant tumors across patients. Despite similar response rates, MHC-II reliant responses were associated with significantly longer durable clinical benefit (Discovery: Median OS=63.6 vs. 34.5 months P=0.0074; Validation: Median OS=37.5 vs. 33.1 months, P=0.040). Characteristics of the tumor immune microenvironment reflected MHC neoantigen reliance, and analysis of immune checkpoints revealed LAG3 as a potential target in MHC-II but not MHC-I reliant responses. This study highlights the value of interpretable machine learning models in elucidating the biological basis of therapy responses. Statement of SignificanceImmune checkpoint blockade works only in a fraction of patients for reasons that are still not fully understood. Our study reveals heterogeneity in the immune responses of ICB responders that correlates with characteristics of the neoantigen landscape. This heterogeneity is accompanied by differences in the duration of clinical benefit as well as by differences as to which immune checkpoint gene serves as a biomarker of ICB response. These findings suggest possible new strategies for improving ICB responses. HighlightsO_LIWe used machine learning to study ICB response across 708 patients from 8 studies across 3 tumor types (melanoma, RCC, and NSCLC). C_LIO_LICombining germline and somatic features improves prediction of ICB response C_LIO_LIInteractions between germline and somatic features reveal mechanisms contributing to ICB sensitivity. C_LIO_LIMHC-I vs. MHC-II reliance implicates LAG3 as a prognostic biomarker in the context of CD4 T cell driven responses. C_LIO_LIMHC-II neoantigen reliant responses provide superior durable clinical benefit in response to ICB. C_LI

cancer biology↗

bioRxiv2023

Predicting neoadjuvant treatment response in triple-negative breast cancer using machine learning

BackgroundNeoadjuvant chemotherapy (NAC) is the standard treatment for early-stage triple negative breast cancer (TNBC). The primary endpoint of NAC is a pathological complete response (pCR). NAC results in pCR in only 30%-40% of TNBC patients. Tumor-infiltrating lymphocytes (TILs), Ki67 and phosphohistone H3 (pH3) are a few known biomarkers to predict NAC response. Currently, systematic evaluation of the combined value of these biomarkers in predicting NAC response is lacking. In this study, the predictive value of markers derived from H&E and IHC stained biopsy tissue was comprehensively evaluated using a supervised machine learning (ML)-based approach. Identifying predictive biomarkers could help guide therapeutic decisions by enabling precise stratification of TNBC patients into responders and partial or non-responders. MethodsSerial sections from core needle biopsies (n=76) were stained with H&E, and immunohistochemically for the Ki67 and pH3 markers, followed by whole slide image (WSI) generation. The resulting WSI triplets were co-registered with H&E WSIs serving as the reference. Separate mask region-based CNN (MRCNN) models were trained with annotated H&E, Ki67 and pH3 images for detecting tumor cells, stromal and intratumoral TILs (sTILs and tTILs), Ki67+, and pH3+ cells. Top image patches with a high density of cells of interest were identified as hotspots. Best classifiers for NAC response prediction were identified by training multiple ML models, and evaluating their performance by accuracy, area under curve, and confusion matrix analyses. ResultsHighest prediction accuracy was achieved when hotspot regions were identified by tTIL counts and each hotspot was represented by measures of tTILs, sTILs, tumor cells, Ki67+, and pH3+ features. Regardless of the hotspot selection metric, a complementary use of multiple histological features (tTILs, sTILs) and molecular biomarkers (Ki67 and pH3) resulted in top ranked performance at the patient level. ConclusionsOverall, our results emphasize that prediction models for NAC response should be based on biomarkers in combination rather than in isolation. Our study provides compelling evidence to support the use of ML-based models to predict NAC response in patients with TNBC.

cancer biology↗

bioRxiv2023

nELISA: A high-throughput, high-plex platform enables quantitative profiling of the secretome

We present the nELISA, a high-throughput, high-fidelity, and high-plex protein profiling platform. DNA oligonucleotides are used to pre-assemble antibody pairs on spectrally encoded microparticles and perform displacement-mediated detection. Spatial separation between non-cognate antibodies prevents the rise of reagent-driven cross-reactivity, while read-out is performed cost-efficiently and at high-throughput using flow cytometry. nELISA can measure both protein concentration and their post-translational modifications. We assembled an inflammatory panel of 191 targets that were multiplexed without cross-reactivity nor impact on performance vs 1-plex signals, with sensitivities as low as 0.1 pg/mL and measurements spanning 7 orders of magnitude. We then performed a large-scale inflammatory-secretome perturbation screen of peripheral blood mononuclear cells (PBMCs), with cytokines as both perturbagens and read-outs, measuring 7,392 samples and generating [~]1.4M protein data points in under a week; a significant advance in throughput compared to other highly multiplexed immunoassays. We uncovered 447 significant cytokine responses, including multiple putatively novel ones, that were conserved across donors and stimulation conditions. We validate nELISA for phenotypic screening, where its capacity to faithfully report hundreds of proteins make it a powerful tool across multiple stages of drug discovery.

bioengineering↗

bioRxiv2022

Structural basis for BIRC6 to balance apoptosis and autophagy

Caspase-9 is the initiator caspase for the intrinsic apoptotic cell death pathway, and is critical to the activation of effector caspases during apoptosis, but how its levels and activities are maintained remains unclear. The gigantic Inhibitor of Apoptosis Protein (IAP) BIRC6/BRUCE/Apollon not only inhibits apoptosis, but also promotes ubiquitination of the key autophagic protein LC3 and inhibits autophagy. Here we show that BIRC6 forms an anti-parallel U-shaped dimer in a 3.6-[A] cryo-EM structure with multiple previously unannotated domains, including a ubiquitin-like domain, and discover that the mitochondria-derived pro-apoptotic factor Smac/DIABLO binds BIRC6 by interacting with one BIR domain, two carbohydrate-binding modules and two helices in the central cavity. Notably, Smac outcompetes the effector caspase 3 and the pro-apoptotic protease HtrA2, but not caspase 9, for binding BIRC6. BIRC6 strongly inhibits cellular activity of caspase 9, but weakly suppresses that of caspase 3. Meanwhile, BIRC6 binds LC3 through an LC3-interacting region, probably following dimer disruption of this BIRC6 region. Deficiency in LC3 ubiquitination promotes autophagy and autophagic degradation of BIRC6, and inhibits apoptosis. Moreover, induction of autophagy promotes autophagic degradation of both procaspase-9 and active caspase-9, but not of effector caspases. These results are important to understand how the balance between apoptosis and autophagy is regulated under pathophysiological conditions.

cell biology↗

bioRxiv2022

Real-Time Spatial Registration for 3D Human Atlas

AO_SCPLOWBSTRACTC_SCPLOWThe human body is made up of about 37 trillion cells (adults). Each cell has its own unique role and is affected by its neighboring cells and environment. The NIH Human BioMolecular Atlas Program (HuBMAP) aims at developing a 3D atlas of human body consisting of organs, vessels, tissues to singe cells with all 3D spatially registered in a single 3D human atlas using tissues obtained from normal individuals across a wide range of ages. A critical step of building the atlas is to register 3D tissue blocks in real-time to the right location of a human organ, which itself consists of complex 3D substructures. The complexity of the 3D organ model, e.g., 35 meshes for a typical kidney, poses a significant computational challenge for the registration. In this paper, we propose a comprehensive framework TICKET (TIssue bloCK rEgisTration) for tissue block registration for 3D human atlas, including (1) 3D mesh pre-processing, (2) spatial queries on intersection relationship and (3) intersection volume computation between organs and tissue blocks. To minimize search space and computation cost, we develop multi-level indexing on both the anatomical structure level and mesh level, and utilize OpenMP for parallel computing. Considering cuboid based shape of the tissue block, we propose an efficient voxelization-based method to estimate the intersection volume. Our experiments demonstrate that the proposed framework is practical and efficient. TICKET is now integrated into the HuBMAP CCF registration portal [1].

bioinformatics↗

bioRxiv2022

Artificial Intelligence based Liver Portal Tract Region Identification and Quantification with Transplant Biopsy Whole-Slide Images

Liver fibrosis staging is clinically important for liver disease progression prediction. As the portal tract fibrotic quantity and size in a liver biopsy correlate with the fibrosis stage, an accurate analysis of portal tract regions is clinically critical. Manual annotations of portal tract regions, however, are time-consuming and subject to large inter- and intra-observer variability. To address such a challenge, we develop a Multiple Up-sampling and Spatial Attention guided UNet model (MUSA-UNet) to segment liver portal tract regions in whole-slide images of liver tissue slides. To enhance the segmentation performance, we propose to use depth-wise separable convolution, the spatial attention mechanism, the residual connection, and multiple up-sampling paths in the developed model. This study includes 53 histopathology whole slide images from patients who received liver transplantation. In total, 6,012 patches derived from 30 images are used for our deep learning model training and validation. The remaining 23 whole slide images are utilized for the model testing. The average liver portal tract segmentation performance of the developed MUSA-UNet is 0.94 (Precision), 0.85 (Recall), 0.89 (F1 Score), 0.89 (Accuracy), 0.80 (Jaccard Index), and 0.91 (Fowlkes-Mallows Index), respectively. The clinical Scheuer fibrosis stage presents a strong correlation with the resulting average portal tract fibrotic area (R=0.681, p<0.001) and portal tract percentage (R=0.335, p=0.02) computed from the MUSA-UNet segmentation results. In conclusion, our developed deep learning model MUSA-UNet can accurately segment portal tract regions from whole-slide images of liver tissue biopsies, presenting its promising potential to assist liver disease diagnosis in a computational manner.

bioinformatics↗

bioRxiv2022

Virtual Liver Needle Biopsy from Reconstructed Three-Dimensional Histopathological Images: Quantification of Sampling Error

IntroductionPrevalently considered as the "gold-standard" for diagnosis of hepatic fibrosis and cirrhosis, the clinical liver needle biopsy is known to be subject to inadequate sampling and a high mis-sampling rate. However, quantifying such sampling bias has been difficult as generating a large number of needle biopsies from the same living patient is practically infeasible. We construct a three-dimension (3D) virtual liver tissue volume by spatially registered high resolution Whole Slide Images (WSIs) of serial liver tissue sections with a novel dynamic registration method. We further develop a Virtual Needle Biopsy Sampling (VNBS) method that mimics the needle biopsy sampling process. We apply the VNBS method to the reconstructed digital liver volume at different tissue locations and angles. Additionally, we quantify Collagen Proportionate Area (CPA) in all resulting virtual needle biopsies in 2D and 3D. ResultsThe staging score of the center 2D longitudinal image plane from each 3D biopsy is used as the biopsy staging score, and the highest staging score of all sampled needle biopsies is the diagnostic staging score. The Mean Absolute Difference (MAD) in reference to the Scheuer and Ishak diagnostic staging scores are 0.22 and 1.00, respectively. The absolute Scheuer staging score difference in 22.22% of sampled biopsies is 1. By the Ishak staging method, 55.56% and 22.22% of sampled biopsies present score difference 1 and 2, respectively. There are 4 (Scheuer) and 6 (Ishak) out of 18 3D virtual needle biopsies with intra-needle variations. Additionally, we find a positive correlation between CPA and fibrosis stages by Scheuer but not Ishak method. Overall, CPA measures suffer large intra- and inter-needle variations. ConclusionsThe developed virtual liver needle biopsy sampling pipeline provides a computational avenue for investigating needle biopsy sampling bias with 3D virtual tissue volumes. This method can be applied to other tissue-based disease diagnoses where the needle biopsy sampling bias substantially affects the diagnostic results.

bioengineering↗

bioRxiv2022

An integrative web-based software tool for multi-dimensional pathology whole-slide image analytics

In the era of precision medicine, human tumor atlas oriented studies have been significantly facilitated by high-resolution, multi-modal tissue based microscopic pathology image analytics. To better support such tissue-based investigations, we develop Digital Pathology Laboratory (DPLab), a publicly available web-based platform, to assist biomedical research groups, non-technical end users, and clinicians for pathology Whole-Slide Image (WSI) visualization, annotation, analysis, and sharing via web browsers. A major advance of this work is the easy-to-follow methods to reconstruct three-dimension (3D) tissue image volumes by registering two-dimension (2D) whole-slide pathology images of serial tissue sections stained by hematoxylin and eosin (H&E), and immunohistochemistry (IHC). The integration of these serial slides stained by different methods provides cellular phenotype and pathophysiologic states in the context of a 3D tissue micro-environment. DPLab is hosted on a publicly accessible server and connected to a backend computational cluster for intensive image analysis computations, with results visualized, downloaded, and shared via a web interface. Equipped with an analysis toolbox of numerous image processing algorithms, DPLab supports continued integration of community-contributed algorithms and presents an effective solution to improve the accessibility and dissemination of image analysis algorithms by research communities. It represents the first step in making next generation tissue investigation tools widely available to the research community, enabling and facilitating discovery of clinically relevant disease mechanisms in a digital 3D tissue space.

bioinformatics↗

bioRxiv2022

Deep Learning Based Registration of Serial Whole-slide Histopathology Images in Different Stains

For routine pathology diagnosis and imaging based biomedical research, Whole-Slide Image (WSI) analyses have been largely limited to a 2D tissue image space. For a more definitive tissue representation to support fine-resolution spatial and integrative analyses, it is critical to extend such tissue based investigations to a 3D tissue space with spatially aligned serial tissue WSIs in different stains, such as Hematoxylin and Eosin (H&E) and Immunohistochemistry (IHC) biomarkers. However, such WSI registration is technically challenged by the overwhelming image scale, the complex histology structure change, and the significant difference in tissue appearances in different stains. We propose a novel translation based deep learning registration network CycGANRegNet that spatially aligns serial WSIs stained in H&E and by IHC biomarkers without prior deformation information for the model training. First, synthetic IHC images are produced from H&E slides through a robust image synthesis algorithm. Next, the synthetic and the real IHC images are registered through a Fully Convolutional Network with multi-scaled deformable vector fields and a joint loss optimization. We perform the registration at the full image resolution, retaining the tissue details in the results. Evaluated with a dataset of 76 breast cancer patients with one H&E and two IHC serial WSIs for each patient, CycGANRegNet outperforms multiple state-of-the-art deep learning based and conventional pathology image registration methods. Our results suggest that CycGAN-RegNet can produce promising registration results with serial WSIs in different stains, enabling integrative 3D tissue-based biomedical investigations.

bioinformatics↗

bioRxiv2022

A Spatial Attention Guided Deep Learning System for Prediction of Pathological Complete Response Using Breast Cancer Histopathology Images

Predicting pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in triple-negative breast cancer (TNBC) patients accurately is direly needed for clinical decision making. pCR is also regarded as a strong predictor of overall survival. In this work, we propose a deep learning system to predict pCR to NAC based on serial pathology images stained with hematoxylin and eosin (H&E) and two immunohistochemical biomarkers (Ki67 and PHH3). To support human prior domain knowledge based guidance and enhance interpretability of the deep learning system, we introduce a human knowledge derived spatial attention mechanism to inform deep learning models of informative tissue areas of interest. For each patient, three serial breast tumor tissue sections from biopsy blocks were sectioned, stained in three different stains, and integrated. The resulting comprehensive attention information from the image triplets is used to guide our prediction system for prognostic tissue regions. The experimental dataset consists of 26,419 pathology image patches of 1,000 x 1,000 pixels from 73 TNBC patients treated with NAC. Image patches from randomly selected 43 patients are used as a training dataset and images patches from the rest 30 are used as a testing dataset. By the maximum voting from patch-level results, our proposed model achieves a 93% patient-level accuracy, outperforming baselines and other state-of-the-art systems, suggesting its high potential for clinical decision making. The codes, the documentation, and example data are available on an open source at: https://github.com/jkonglab/PCR_Prediction_Serial_WSIs_biomarkers

bioinformatics↗

bioRxiv2022

MultiHeadGAN: A Deep Learning Method for Low Contrast Retinal Pigment Epithelium Cells Segmentation in Fluorescent Flatmount Microscopy Images

BackgroundRetinal pigment epithelium (RPE) aging is an important cause of vision loss. As RPE aging is accompanied by changes in cell morphological features, an accurate segmentation of RPE cells is a prerequisite to such morphology analyses. Due the overwhelmingly large cell number, manual annotations of RPE cell borders are time-consuming. Computer based methods do not work well on cells with weak or missing borders in the impaired RPE sheet regions. MethodTo address such a challenge, we develop a semi-supervised deep learning approach, namely MultiHeadGAN, to segment low contrast cells from impaired regions in RPE flatmount images. The developed deep learning model has a multi-head structure that allows model training with only a small scale of human annotated data. To strengthen model learning effect, we further train our model with RPE cells without ground truth cell borders by generative adversarial networks. Additionally, we develop a new shape loss to guide the network to produce closed cell borders in the segmentation results. ResultsIn this study, 155 annotated and 1,640 unlabeled image patches are included for model training. The testing dataset consists of 200 image patches presenting large impaired RPE regions. The average RPE segmentation performance of the developed model MultiHeadGAN is 85.4 (correct rate), 88.8 (weighted correct rate), 87.3 (precision), and 80.1 (recall), respectively. Compared with other state-of-the-art deep learning approaches, our method demonstrates superior qualitative and quantitative performance. ConclusionsSuggested by our extensive experiments, our developed deep learning method can accurately segment cells from RPE flatmount microscopy images and is promising to support large scale cell morphological analyses for RPE aging investigations.

bioinformatics↗

bioRxiv2021

Fast anther dehiscence state recognition system establishing by deep learning to screen heat tolerant cotton

Cotton is one of the most economically important crops in the world. The fertility of male reproductive organs is a key determinant of cotton yield. The anther dehiscence or indehiscence directly determine the probability of fertilization in cotton. Thus, the rapid and accurate identification of cotton anther dehiscence status is important for judging anther growth status and promoting genetic breeding research. The development of computer vision technology and the advent of big data have prompted the application of deep learning techniques to agricultural phenotype research. Therefore, two deep learning models (Faster R-CNN and YOLOv5) were proposed to detect the number and dehiscence status of anthers. The single-stage model based on YOLOv5 has higher recognition efficiency and the ability to deploy to the mobile end. Breeding researchers can apply this model to terminals to achieve a more intuitive understanding of cotton anther dehiscence status. Moreover, three improvement strategies of Faster R-CNN model were proposed, the improved model has higher detection accuracy than YOLOv5 model. In addition, the percentage of dehiscent anther of randomly selected 30 cotton varieties were observed from cotton population under normal temperature and high temperature (HT) conditions through the integrated Faster R-CNN model and manual observation. The result showed HT varying decreased the percentage of dehiscent anther in different cotton lines, consistent with the manual method. Thus, this system can help us to rapid and accurate identification of HT-tolerant cotton. One sentence summaryThe deep learning technique was applied to identify the anther dehiscence state for the first time to quickly screen heat tolerant cotton varieties and help to explore key genetic improvement genes.

bioengineering↗