Biology subjects

Du, K.

Publications and source records attributed to Du, K..

At least 55 records · Page 3Linked to original sources

bioRxiv2022

Understanding the Dynamics of Fluid-Structure Interaction with an Air Deflected Microfluidic Chip (ADMC)

A deformable microfluidic system and a fluidic dynamic model have been successfully coupled to understand the dynamic fluid-structure interaction in transient flow, designed to understand the dentine hypersensitivity caused by hydrodynamic theory. The Polydimethylsiloxane thin sidewalls of the microfluidic chip are deformed with air pressure ranging from 50 to 500 mbar to move the liquid meniscus in the central liquid channel. The displacement is recorded and compared with our new theoretical model derived from the unsteady Bernoulli equation. We show that our theoretical model can well predict the ending point of the liquid displacement as well as the dynamics process, regardless of the wall thickness. Moreover, an overshooting and oscillation phenomenon is observed by reducing the friction factor by a few orders which could be the key to explain the dentine hypersensitivity caused by the liquid movement in the dentine tubules.

bioengineering↗

bioRxiv2022

A GPU-based computational framework that bridges Neuron simulation and Artificial Intelligence

Biophysically detailed multi-compartment models are powerful tools to explore computational principles of the brain and also serve as a theoretical framework to generate algorithms for artificial intelligence (AI) systems. However, the expensive computational cost severely limits the applications in both the neuroscience and AI fields. The major bottleneck during simulating detailed compartment models is the ability of a simulator to solve large systems of linear equations. Here, we present a novel Dendritic Hierarchical Scheduling (DHS) method to markedly accelerate such process. We theoretically prove that the DHS implementation is computationally optimal and accurate. This GPU-based method performs at 2-3 orders of magnitude higher speed than that of the classic serial Hines method in the conventional CPU platform. We build a DeepDendrite framework, which integrates the DHS method and the GPU computing engine of the NEURON simulator and demonstrate applications of DeepDendrite in neuroscience and AI tasks. We investigated how spatial patterns of spine inputs affect neuronal excitability in a detailed human pyramidal neuron model with 25,000 spines; and examined how dendrites protect morphologically detailed neural networks against adversarial attacks in typical image classification tasks.

neuroscience↗

bioRxiv2022

Retina Gap Junction Networks Facilitate Blind Denoising in Visual Hierarchy

Gap junctions in the retina are electrical synapses, which strength is regulated byambient light conditions. Such tunable synapses are crucial for the denoising function of the early visual system. However, it is unclear that how the plastic gap junction network processes unknown noise, specifically how this process works synergistically with the brains higher visual centers. Inspired by the electrically coupled photoreceptors, we develop a computational model of the gap junction filter (G-filter). We show that G-filter is an effective blind denoiser that converts different noise distributions into a similar form. Next, since deep convolutional neural networks (DCNNs) functionally reflect some intrinsic features of the visual cortex, we combine G-filter with DCNNs as retina and ventral visual pathways to investigate the relationship between retinal denoising processing and the brains high-level functions. In the image denoising and reconstruction task, G-filter dramatically improve the classic deep denoising convolutional neural network (DnCNN)s ability to process blind noise. Further, we find that the gap junction strength of the G-filter modulates the receptive field of DnCNNs output neurons by the Integrated Gradients method. At last, in the image classification task, G-filter strengthens the defense of state-of-the-arts DCNNs (ResNet50, VGG19 and InceptionV3) against blind noise attacks, far exceeding human performance when noise is large. Our results indicate G-filter significantly enhance DCNNs ability on various blind denoising tasks, implying an essential role for retina gap junction networks in high-level visual processing.

neuroscience↗

bioRxiv2021

Amplification-free detection of viral RNA by super resolution imaging-based CRISPR/Cas13a System

RNA detection is crucial for biological research and clinical diagnosis. The current methods include both direct and amplification-based RNA detection. These methods require complicated procedures, suffering from low sensitivity, slow turnaround, and amplification bias. The CRISPR/Cas13a system is a direct RNA detection method via target RNA induced collateral cleavage activity. However, to detect low concentration RNA with CRISPR/Cas13a, target amplification is always required. Herein, we optimize the components of the CRISPR/Cas13a assay to enhance the sensitivity of viral RNA detection which improve the detection limit from 1 pM up to 100 fM. In addition, the integration of CRISPR/Cas13a biosensing and single molecule super resolution imaging is a novel strategy for direct and amplification-free RNA detection. After surface modification, fluorescent RNA reporters are immobilized on the glass coverslip surface and fluorescent signals are captured by total internal reflection fluorescence microscopy (TIRFM), shifting the measurement from spectroscopy to imaging. We quantify the fluorescence signal intensity before and after collateral cleavage of the CRISPR system when viral RNA is present and achieve a detection limit of 10 fM. Therefore, we provide a novel TIRFM-based system to visualize the CRISPR trans-cleavage for direct and robust RNA detection.

bioengineering↗

bioRxiv2021

Coupled protein synthesis and ribosome-guided piRNA processing on mRNAs

PIWI-interacting small RNAs (piRNAs) protect the germline genome and are essential for fertility. Previously, we showed that ribosomes guide the biogenesis of piRNAs from long non-coding RNAs (lncRNAs) after translating the short open reading frames (ORFs) near their 5 cap. It remained unclear, however, how ribosomes proceed downstream of ORFs and how piRNA precursors distinguish from other RNAs. It is thus important to test whether a short ORF length is required for substrate recognition for ribosome guided-piRNA biogenesis. Here, we characterized a poorly understood class of piRNAs that originate from the 3 untranslated regions (3UTRs) of protein coding genes in mice and chickens. We demonstrate that their precursors are full-length mRNAs and that post-termination 80S ribosomes guide piRNA production on 3UTRs after translation of upstream long ORFs. Similar to non-sense mediated decay (NMD), piRNA biogenesis degrades mRNA right after pioneer rounds of translation and fine-tunes protein production from mRNAs. Interestingly, however, we found that NMD, along with other surveillance pathways for ribosome recycling are temporally sequestered during the pachytene stage to allow for robust piRNA production. Although 3UTR piRNA precursor mRNAs code for distinct proteins in mice and chickens, they all harbor embedded transposable elements (TEs) and produce piRNAs that cleave TEs, suggesting that TE suppression, rather than the function of proteins, is the primary evolutionary force maintaining a subset of mRNAs as piRNA precursors. Altogether, we discover a function of the piRNA pathway in fine-tuning protein production and reveal a conserved, general piRNA biogenesis mechanism that recognizes translating RNAs regardless of their ORF length in amniotes.

molecular biology↗

bioRxiv2021

U-RISC: An Ultra-high Resolution EM Dataset Challenging Existing Deep Learning Algorithms

Connectomics is a developing field aiming at reconstructing the connection of the neural system at nanometer scale. Computer vision technology, especially deep learning methods used in image processing, has promoted connectomic data analysis to a new era. However, the performance of the state-of-the-art methods still falls behind the demand of scientific research. Inspired by the success of ImageNet, we present the U-RISC, an annotated Ultra-high Resolution Image Segmentation dataset for Cell membrane, which is the largest cell membrane annotated Electron Microscopy (EM) dataset with a resolution of 2.18nm/pixel. Multiple iterative annotations ensured the quality of the dataset. Through an open competition, we reveal that the performance of current deep learning methods still has a considerable gap with human-level, different from ISBI 2012, on which the performance of deep learning is close to human. To explore the causes of this discrepancy, we analyze the neural networks with a visualization method, attribution analysis. We find that in U-RISC, it requires a larger area around a pixel to predict whether the pixel belongs to the cell membrane or not. Finally, we integrate currently available methods to provide a new benchmark (0.67, 10% higher than the leader of competition, 0.61) for cell membrane segmentation on U-RISC and propose some suggestions in developing deep learning algorithms. The U-RISC dataset and the deep learning codes used in this paper will be publicly available.

neuroscience↗

bioRxiv2021

Understanding Microbeads Stacking in Deformable Nano-Sieve for Efficient Plasma Separation and Blood Cell Retrieval

Efficient separation of blood cells and plasma is key for numerous molecular diagnosis and therapeutics applications. Despite various microfluidics-based separation strategies have been developed, a simple, reliable, and multiplexing separation device that can process a large volume of blood is still missing. Here we show a microbead packed deformable microfluidic system that can efficiently separate highly purified plasma from whole blood as well as retrieve blocked blood cells from the device. Combining microscope imaging, optical tomography scanning, and computational fluidic modeling, a highly accurate model is constructed to understand the link between the mechanical properties of the microfluidics, flow rate, and microbeads packing/leaking, which supports and rationalizes the experimental observations. This deformable nano-sieve device establishes a key technology for centrifuge-free diagnosis and treatment of bloodborne diseases and may be important for the design of next-generation deformable microfluidics for separation applications.

bioengineering↗

bioRxiv2021

A General LSTM-based Deep Learning Method for Estimating Neuronal Models and Inferring Neural Circuitry

Computational neural models are essential tools for neuroscientists to study the functional roles of single neurons or neural circuits. With the recent advances in experimental techniques, there is a growing demand to build up neural models at single neuron or large-scale circuit levels. A long-standing challenge to build up such models lies in tuning the free parameters of the models to closely reproduce experimental recordings. There are many advanced machine-learning-based methods developed recently for parameter tuning, but many of them are task-specific or requires onerous manual interference. There lacks a general and fully-automated method since now. Here, we present a Long Short-Term Memory (LSTM)-based deep learning method, General Neural Estimator (GNE), to fully automate the parameter tuning procedure, which can be directly applied to both single neuronal models and large-scale neural circuits. We made comprehensive comparisons with many advanced methods, and GNE showed outstanding performance on both synthesized data and experimental data. Finally, we proposed a roadmap centered on GNE to help guide neuroscientists to computationally reconstruct single neurons and neural circuits, which might inspire future brain reconstruction techniques and corresponding experimental design. The code of our work will be publicly available upon acceptance of this paper.

neuroscience↗

bioRxiv2020

Quantitative analysis of Histone modifications in gene silencing

Gene silencing in budding yeast is mediated by Sir protein binding to unacetylated nucleosomes to form a chromatin structure that inhibits transcription. This transcriptional silencing is characterized by the high-fidelity transmission of the silent state. Despite its relative stability, the constituent parts of the silent state are in constant flux giving rise to a model that silent loci can tolerate such fluctuations without functional consequences. However, the level of tolerance is unknown and we developed a method to measure the threshold of histone acetylation that causes the silent chromatin state to switch to the active state. We show that loss of silencing required between 50% and 75% of the unacetylated histones to be replaced with acetylated histone mimics. The precise levels of unacetylated nucleosomes required varied from locus to locus and was influenced by both silencer strength and UAS enhancer/promoter strength. Simple calculations suggest that an approximately 50% reduction in the ability of acetylases to acetylate individual nucleosomes across a large domain may be sufficient to generate a transcriptionally silent region in the nucleus.

molecular biology↗

bioRxiv2020

Collagen induced inflammations

Collagen-induced arthritis (CIA) mouse model is currently the most widely used and reliable autoimmune model to study rheumatoid arthritis. In this model, we used bovine type II collagen (CII) and complete Freunds adjuvant (CFA) or incomplete Freunds adjuvant (IFA) to form emulsifier, and mice were injected intradermal to induce autoimmune arthritis (CIA). In this model, we ground bovine collagen type II (CII) with complete Freunds adjuvant (CFA) or incomplete Freunds adjuvant (IFA) to form emulsifiers, and intradermal injection in mice induced autoimmune arthritis (CIA). This article describes the mouse, CFA strains, key emulsification, anesthesia, and injection immune techniques, as well as the incidence, date of onset, score, pathological results of arthritis. The total time for preparation of reagents and immunization of 20 mice was about 2-2.5 hours. In this protocol, we induced a high incidence of CIA with DBA/1J in genetically susceptible mice and assessed the severity and pathology of the disease, at the same time we found that CII also can induced enteritis, including ileitis and colitis. The initial symptoms of arthritis appear in the 24-26 days of the experiment, that is, 3-5 days after the second immunization, the peak period of inflammation was 30-36 days, the arthritis incidence about 90-100%, at the same time, the incidence of enteritis and arthritis were the same, small intestinal inflammation was more severe, but the duration was short; while the colonic inflammation was mild, and the duration was longer than enteritis, we named it collagen induced inflammations (CIIs).

immunology↗

bioRxiv2020

A 180 My-old female-specific genome region in sturgeon reveals the oldest known vertebrate sex determining system with undifferentiated sex chromosomes

Several hypotheses explain the prevalence of undifferentiated sex chromosomes in poikilothermic vertebrates. Turnovers change the master sex determination gene, the sex chromosome or the sex determination system (e.g. XY to WZ). Jumping master genes stay main triggers but translocate to other chromosomes. Occasional recombination (e.g. in sex-reversed females) prevents sex chromosome degeneration. Recent research has uncovered conserved heteromorphic or even homomorphic sex chromosomes in several clades of non-avian and non-mammalian vertebrates. Sex determination in sturgeons (Acipenseridae) has been a long-standing basic biological question, linked to economical demands by the caviar-producing aquaculture. Here, we report the discovery of a sex-specific sequence from sterlet (Acipenser ruthenus). Using chromosome-scale assemblies and pool-sequencing, we first identified a ~16 kb female-specific region. We developed a PCR-genotyping test, yielding female-specific products in six species, spanning the entire phylogeny with the most divergent extant lineages (A. sturio, A. oxyrinchus vs. A. ruthenus, Huso huso), stemming from an ancient tetraploidization. Similar results were obtained in two octoploid species (A. gueldenstaedtii, A. baerii). Conservation of a female-specific sequence for a long period, representing 180 My of sturgeon evolution, and across at least one polyploidization event, raises many interesting biological questions. We discuss a conserved undifferentiated sex chromosome system with a ZZ/ZW-mode of sex determination and potential alternatives.

evolutionary biology↗

bioRxiv2020

High contrast probe cleavage detection on porous silicon biosensors via quantum dot labeled DNA probes

Using porous silicon (PSi) interferometer sensors, we show the first experimental implementation of the high contrast probe cleavage detection (HCPCD) mechanism. HCPCD makes use of dramatic optical signal amplification caused by cleavage of high-contrast nanoparticle labels on probes instead of the capture of low-index biological molecules. An approximately 2 nm reflectance peak shift was detected after cleavage of DNA-quantum dot probes from the PSi surface via exposure to a 12.5 nM DNase enzyme solution for 2 hrs. This signal change is 20 times greater than the resolution of the spectrometer used for the interferometric measurements, and the interferometric measurements agree with the interferometric response predicted by simulations and fluorescence measurements. These proof of principle experiments show a clear path to real-time, highly sensitive and inexpensive point-of-care readout for a broad range of biological diagnostic assays that generate signal via nucleic acid cleavage.

bioengineering↗

bioRxiv2020

The developmental and genetic architecture of the sexually selected male ornament of swordtails

Sexual selection results in sex-specific characters like the conspicuously pigmented extension of the ventral tip of the caudal fin - the "sword" - in males of several species of Xiphophorus fishes. To uncover the genetic architecture underlying sword formation and to identify genes that are associated with its development, we characterized the sword transcriptional profile and combined it with genetic mapping approaches. Results showed that the male ornament of swordtails develops from a sexually non-dimorphic prepattern of transcription factors in the caudal fin. Among genes that constitute the exclusive sword transcriptome only two are located in the genomic region associated with this trait; the chaperone, fkbp9, and the potassium channel, kcnh8 that in addition to its neural function performs a role known to affect fin growth. This indicates that during evolution of swordtails a brain gene has been recruited for an additional function in establishing a male ornament.

evolutionary biology↗

bioRxiv2020

A Molecular Mechanism for Probabilistic Bet-hedging and its Role in Viral Latency

Probabilistic bet hedging, a strategy to maximize fitness in unpredictable environments by matching phenotypic variability to environmental variability, is theorized to account for the evolution of various fate-specification decisions, including viral latency. However, the molecular mechanisms underlying bet-hedging remain unclear. Here, we report that large variability in protein abundance within individual herpesvirus virion particles enables probabilistic bet hedging between viral replication and latency. Super-resolution imaging of individual virions of the human herpesvirus cytomegalovirus (CMV) showed that virion-to-virion levels of pp71 tegument protein--the major viral transactivator protein--exhibit extreme variability. This super-Poissonian tegument variability promoted alternate replicative strategies: high virion pp71 levels enhance viral replicative fitness but, strikingly, impede silencing, whereas low virion pp71 levels reduce fitness but promote silencing. Overall, the results indicate that stochastic tegument packaging provides a mechanism enabling probabilistic bet hedging between viral replication and latency. SIGNIFICANCEProbabilistic bet hedging is a generalized diversification strategy to maximize fitness in unpredictable environments, and has been proposed as an evolutionary basis for herpesvirus latency. However, the molecular mechanisms enabling probabilistic bet hedging have remained elusive. Here, we find that the human herpesvirus cytomegalovirus--a major cause of birth defects and transplant failures--utilizes stochastic variability in the abundance of a protein packaged into individual viral particles to enable probabilistic bet hedging between alternate viral states.

molecular biology↗

bioRxiv2020

RADSex: a computational workflow to study sex determination using Restriction Site-Associated DNA Sequencing data

The study of sex determination and sex chromosome organisation in non-model species has long been technically challenging, but new sequencing methodologies are now enabling precise and high-throughput identification of sex-specific genomic sequences. In particular, Restriction Site-Associated DNA Sequencing (RAD-Seq) is being extensively applied to explore sex determination systems in many plant and animal species. However, software designed to specifically search for sex-biased markers using RAD-Seq data is lacking. Here, we present RADSex, a computational analysis workflow designed to study the genetic basis of sex determination using RAD-Seq data. RADSex is simple to use, requires few computational resources, makes no prior assumptions about type of sex-determination system or structure of the sex locus, and offers convenient visualization through a dedicated R package. To demonstrate the functionality of RADSex, we re-analyzed a published dataset of Japanese medaka, Oryzias latipes, where we uncovered a previously unknown Y chromosome polymorphism. We then used RADSex to analyze new RAD-Seq datasets from 15 fish species spanning multiple systematic orders. We identified the sex determination system and sex-specific markers in six of these species, five of which had no known sex-markers prior to this study. We show that RADSex greatly facilitates the study of sex determination systems in non-model species and outperforms the commonly used RAD-Seq analysis software STACKS. RADSex in speed, resource usage, ease of application, and visualization options. Furthermore, our analysis of new datasets from 15 species provides new insights on sex determination in fish.

bioinformatics↗

bioRxiv2020

Magnetic Bead-Quantum Dot (MB-Qdot) CRISPR Assay for Instrument-Free Viral DNA Detection

We have developed a novel detection system which couples CRISPR-Cas recognition of target sequences, Cas mediated nucleic acid probe cleavage, and quantum dots as highly sensitive reporter molecules for instrument-free detection of viral nucleic acid targets. After target recognition and Cas mediated cleavage of biotinylated ssDNA probe molecules, the probe molecules are bound to magnetic particles. A complimentary ssDNA oligonucleotide quantum dot conjugate is then added, which only hybridizes to un-cleaved probes on the magnetic beads. After separation of hybridized from unhybridized quantum dot conjugates by magnetic sequestration, the signal is measured fluorometrically to provide a signal proportional to the cleaved probes and therefore the amount of target nucleic acid. To demonstrate the power of this assay, a 250 bp DNA target sequence matching a portion of the African swine fever virus (ASFV) genome is used and a detection limit of ~0.5 nM is achieved without target amplification using a simple portable ultraviolet flashlight. The positive samples are readily confirmed by visual inspection, completely avoiding the need for complicated devices and instruments. This work establishes the feasibility of a simple, instrument free assay for rapid nucleic acid screening in both hospitals and point-of-care settings.

biochemistry↗

bioRxiv2020

Integrated Micropillar Polydimethylsiloxane Accurate CRISPR Detection (IMPACT) System for Rapid Viral DNA Sensing

A fully Integrated Micropillar Polydimethylsiloxane Accurate CRISPR Detection (IMPACT) system is developed for viral DNA detection. This powerful system is patterned with high-aspect ratio micropillars to enhance reporter probe binding. After surface modification and probe immobilization, CRISPR Cas12a/crRNA complex is injected into the fully enclosed system. With the presence of double-stranded DNA target, the CRISPR enzyme is activated and non-specifically cleaves the ssDNA reporters initially immobilized on the micropillars. This collateral cleavage releases fluorescence dyes into the assay, and the intensity is linearly proportional to the target DNA concentration ranging from 0.1 to 10 nM. Importantly, this system does not rely on traditional dye-quencher labeled probe thus eliminating the fluorescence background presented in the assay. Furthermore, our one-step detection protocol is performed at isothermal conditions (37{degrees}C) without using complicated and time-consuming off-chip probe hybridization and denaturation. This miniaturized and fully packed IMPACT chip demonstrates rapid, sensitive, and simple nucleic acid detection and is an ideal candidate for the next generation molecular diagnostic platform for point-of-care (POC) applications, responding to emerging and deadly pathogen outbreaks.

bioengineering↗

bioRxiv2020

Single Chlamydomonas reinhardi (C. reinhardtii) cell separation from bacterial cells and auto-fluorescence tracking with a nano-sieve device

A planar, transparent, and adaptable nano-sieve device is developed for efficient microalgae/bacteria separation. In our strategy, a sacrificial layer is applied in the dual photolithography patterning to achieve a one-dimensional channel with a very low aspect ratio (1:10,000). Microalgae/bacteria mixture is then introduced into the deformable polydimethylsiloxane (PDMS) nano-channel. The hydrodynamic deformation of the nano-channel is regulated to allow the bacteria cells to pass through while leaving the microalgae cells trapped in the device. At a flow rate of 4 l/min, ~100% of the microalgae cells are trapped in the device. Additionally, this device is capable of immobilizing single cells in a transparent channel for auto-fluorescence tracking. These microalgae cells demonstrate minimal photo-bleaching over 250 s laser exposure and can be used to monitor hazardous compounds in the sample with a continuous flow fashion. Our method will be valuable to purify microalgae samples containing contaminations and study single cell heterogeneity.

bioengineering↗