Biology subjects

Lu, H.

Publications and source records attributed to Lu, H..

At least 217 records · Page 12Linked to original sources

bioRxiv2021

Angle-Tuned Coils: Enabling Building Blocks for High Performance and Multisite TMS Systems

Coordinated whole-brain neural dynamics are essential for proper control of the functionality of different brain systems. Multisite simultaneous or sequential stimulations may provide tools for mechanistic studies of brain functions and the treatment of neuropsychiatric disorders. Conventional circular and figure-8 Transcranial Magnetic Stimulation (TMS) coils occupy a large footprint, and it is difficult to reach desired multiple stimulation locations with close proximity for comprehensive multisite stimulations. These conventional coils, limited by the depth-spread tradeoff rule, also lack the required focality for targeted stimulations. In this work, we propose and demonstrate angle-tuned TMS (AT) coils with an intrinsically reduced footprint with their geometric arrangements of stacking and angle tuning. The stimulation depth can be adjusted with the coil stacking number, and the field spread can be reduced by increasing the tilted wire-wrapping angle of the coils. With either smaller or larger diameter coils than a standard commercial figure-8 coil, we show, theoretically and experimentally, improved field decay rate and field intensity, and the reduced field spread spot size at different stimulation depths. These results indicate that the proposed novel coil establishes a better depth-spread tradeoff curve than the conventional circular and figure-8 coils. This coil design has a simple and single element structure and provides a promising solution for an improved multisite brain stimulation performance and serves as the building block of more complex coils for further depth-spread improvements.

neuroscience↗

bioRxiv2021

Cell transcriptomic atlas of the non-human primate Macaca fascicularis

Studying tissue composition and function in non-human primates (NHP) is crucial to understand the nature of our own species. Here, we present a large-scale single-cell and single-nucleus transcriptomic atlas encompassing over one million cells from 43 tissues from the adult NHP Macaca fascicularis. This dataset provides a vast, carefully annotated, resource to study a species phylogenetically close to humans. As proof of principle, we have reconstructed the cell-cell interaction networks driving Wnt signalling across the body, mapped the distribution of receptors and co-receptors for viruses causing human infectious diseases and intersected our data with human genetic disease orthologous coordinates to identify both expected and unexpected associations. Our Macaca fascicularis cell atlas constitutes an essential reference for future single-cell studies in human and NHP.

cell biology↗

bioRxiv2021

scBERT: a Large-scale Pretrained Deep Langurage Model for Cell Type Annotation of Single-cell RNA-seq Data

Annotating cell types based on the single-cell RNA-seq data is a prerequisite for researches on disease progress and tumor microenvironment. Here we show existing annotation methods typically suffer from lack of curated marker gene lists, improper handling of batch effect, and difficulty in leveraging the latent gene-gene interaction information, impairing their generalization and robustness. We developed a pre-trained deep neural network-based model scBERT (single-cell Bidirectional Encoder Representations from Transformers) to overcome the challenges. Following BERTs approach of pre-train and fine-tune, scBERT obtains a general understanding of gene-gene interaction by being pre-trained on huge amounts of unlabeled scRNA-seq data and is transferred to the cell type annotation task of unseen and user-specific scRNA-seq data for supervised fine-tuning. Extensive and rigorous benchmark studies validated the superior performance of scBERT on cell type annotation, novel cell type discovery, robustness to batch effect, and model interpretability.

bioinformatics↗

bioRxiv2021

High-resolution spatiotemporal transcriptomic maps of developing Drosophila embryos and larvae

Drosophila has long been a successful model organism in multiple fields such as genetics and developmental biology. Drosophila genome is relatively smaller and less redundant, yet largely conserved with mammals, making it a productive model in studies of embryogenesis, cell signaling, disease mechanisms, etc. Spatial gene expression pattern is critical for understanding of complex signaling pathways and cell-cell interactions, whereas temporal gene expression changes need to be tracked during highly dynamic activities such as tissue development and disease progression. Systematic studies in Drosophila as a whole are still impeded by lack of these spatiotemporal transcriptomic information. Drosophila embryos and tissues are of relatively small size, limiting the application of current technologies to comprehensively resolve their spatiotemporal gene expression patterns. Here, utilizing SpaTial Enhanced REsolution Omics-sequencing (Stereo-seq), we dissected the spatiotemporal transcriptomic changes of developing Drosophila with high resolution and sensitivity. Our data recapitulated the spatial transcriptomes of embryonic and larval development in Drosophila. With these data, we identified known and previously undetected subregions in several tissues during development, and revealed known and potential gene regulatory networks of transcription factors within their topographic background. We further demonstrated that Stereo-seq data can be used for 3D reconstruction of Drosophila embryo spatial transcriptomes. Our data provides Drosophila research community with useful resources of spatiotemporally resolved transcriptomic information across developmental stages.

developmental biology↗

bioRxiv2021

Brain functional connectome defines a transdiagnostic dimension associated with cognition and psychopathology in youth

BackgroundCognitive dysfunction and high-order psychopathologic dimensions are two main classes of transdiagnostic factors related to psychiatric disorders. They may link to common or distinct core brain networks underlying developmental risk of psychiatric disorders. MethodThe current study is a longitudinal investigation with 11,875 youths aged 9-to 10-years-old at study onset, from the Adolescent Brain Cognitive Development study. A machine-learning approach based on canonical correlation analysis was used to identify latent dimensional associations of the resting-state functional connectome with multi-domain behavioral assessments of cognitive functions and psychopathological problems. For the latent rsFC factor showing a robust behavioral association, its ability to predict psychiatric disorders was assessed using two-year follow-up data and its genetic association was evaluated using twin data from the same cohort. ResultA latent functional connectome pattern was identified that showed a strong and generalizable association with the multi-domain behavioral assessments (5-fold cross validation: {rho} = 0.68~0.73, for the training set (N = 5096); {rho} = 0.56 ~ 0.58, for the test set (N = 1476)). This functional connectome pattern was highly heritable (h2 = 74.42%, 95% CI: 56.76%-85.42%), exhibited a dose-response relationship with cumulative number of psychiatric disorders assessed concurrently and 2-years post-MRI-scan, and predicted the transition of diagnosis across disorders over the 2-year follow-up period. ConclusionThese findings provide preliminary evidence for a transdiagnostic connectome-based measure that underlies individual differences in developing psychiatric disorders in early adolescence.

neuroscience↗

bioRxiv2021

Deep learning redesign of PETase for practical PET degrading applications

Plastic waste poses an ecological challenge1. While current plastic waste management largely relies on unsustainable, energy-intensive, or even hazardous physicochemical and mechanical processes, enzymatic degradation offers a green and sustainable route for plastic waste recycling2. Poly(ethylene terephthalate) (PET) has been extensively used in packaging and for the manufacture of fabrics and single-used containers, accounting for 12% of global solid waste3. The practical application of PET hydrolases has been hampered by their lack of robustness and the requirement for high processing temperatures. Here, we use a structure-based, deep learning algorithm to engineer an extremely robust and highly active PET hydrolase. Our best resulting mutant (FAST-PETase: Functional, Active, Stable, and Tolerant PETase) exhibits superior PET-hydrolytic activity relative to both wild-type and engineered alternatives, (including a leaf-branch compost cutinase and its mutant4) and possesses enhanced thermostability and pH tolerance. We demonstrate that whole, untreated, post-consumer PET from 51 different plastic products can all be completely degraded by FAST-PETase within one week, and in as little as 24 hours at 50 {degrees}C. Finally, we demonstrate two paths for closed-loop PET recycling and valorization. First, we re-synthesize virgin PET from the monomers recovered after enzymatic depolymerization. Second, we enable in situ microbially-enabled valorization using a Pseudomonas strain together with FAST-PETase to degrade PET and utilize the evolved monomers as a carbon source for growth and polyhydroxyalkanoate production. Collectively, our results demonstrate the substantial improvements enabled by deep learning and a viable route for enzymatic plastic recycling at the industrial scale.

bioengineering↗

bioRxiv2021

An intein-mediated split-nCas9 system for base editing in plants

Virus-assisted delivery of the clustered regularly interspaced short palindromic (CRISPR)/CRISPR-associated (Cas) system represents a promising approach for editing plant genomes. However, the relatively large size of the CRISPR/Cas9 system is challenging to package into viral vectors with confined packaging capacity. To address this technical challenge, we developed a strategy that splits the required CRISPR-Cas9 components across a dual-vector system in which CRISPR-Cas reassembles into an active form following co-infection to achieve targeted genome editing in plant cells. An intein-mediated split system was adapted and optimized in plant cells by successfully demonstrating split-eYGFPuv expression. Using a plant-based biosensor, we demonstrated for the first time that the split-SpnCas9 is capable of inducing efficient base editing in plant cells and identified several valid split sites for future biodesign strategies. Overall, this strategy provides new opportunities to bridge different CRISPR/Cas9 tools including base editor, prime editor, and CRISPR activation with virus-mediated gene editing.

plant biology↗

bioRxiv2021

Plant-based biosensors for detecting CRISPR-mediated genome engineering

CRISPR/Cas has recently emerged as the most reliable system for genome engineering in various species. However, concerns about risks associated with CRISPR/Cas9 technology are increasing on potential unintended DNA changes that might accidentally arise from CRISPR gene editing. Developing a system that can detect and report the presence of active CRIPSR/Cas tools in biological systems is therefore very necessary. Here, we developed the real-time detection systems that can spontaneously indicate CRISPR-Cas tools for genome editing and gene regulation including CRISPR/Cas9 nuclease, base editing, prime editing and CRISPRa in plants. Using the fluorescence-based molecular biosensors, we demonstrated that the activities of CRISPR/Cas9 nuclease, base editing, prime editing and CRIPSRa can be effectively detected in transient expression via protoplast transformation and leaf infiltration (in Arabidopsis, poplar, and tobacco) and stable transformation in Arabidopsis.

plant biology↗

bioRxiv2021

Structure of benzothiadiazine at zwitterionic phospholipid cell membranes

The use of drugs derived from benzothiadiazine, which is a bicyclic heterocyclic benzene derivative, has become a widespread treatment for diseases such as hypertension (treated with diuretics such as bendroflumethiazide or chlorothiazide), low blood sugar (treated with non-diuretic diazoxide) or the human immunodeficiency virus, among others. In this work we have investigated the interactions of benzothiadiazine with the basic components of cell membranes and solvents such as phospholipids, cholesterol, ions and water. The analysis of the mutual microscopic interactions is of central importance to elucidate the local structure of benzothiadiazine as well as the mechanisms responsible for the access of benzothiadiazine to the interior of the cell. We have performed molecular dynamics simulations of benzothiadiazine embedded in three different model zwitterionic bilayer membranes made by dimyristoilphosphatidylcholine, dioleoylphosphatidylcholine, 1,2-dioleoyl-sn-glycero-3-phosphoserine and cholesterol inside aqueous sodium-chloride solution in order to systematically examine microscopic interactions of benzothiadiazine with the cell membrane at liquid-crystalline phase conditions. From data obtained through radial distribution functions, hydrogen-bonding lengths and potentials of mean force based on reversible work calculations, we have observed that benzothiadiazine has a strong affinity to stay at the cell membrane interface although it can be fully solvated by water in short periods of time. Furthermore, benzothiadiazine is able to bind lipids and cholesterol chains by means of single and double hydrogen-bonds of different characteristic lengths.

biophysics↗

bioRxiv2021

Deep learning based kcat prediction enables improved enzyme constrained model reconstruction

Enzyme turnover numbers (kcat values) are key parameters to understand cell metabolism, proteome allocation and physiological diversity, but experimentally measured kcat data are sparse and noisy. Here we provide a deep learning approach to predict kcat values for metabolic enzymes in a high-throughput manner with the input of substrate structures and protein sequences. Our approach can capture kcat changes for mutated enzymes and identify amino acid residues with great impact on kcat values. Furthermore, we applied the approach to predict genome scale kcat values for over 300 yeast species, demonstrating that the predicted kcat values are consistent with current evolutional understanding. Additionally, we designed an automatic pipeline using the predicted kcat values to parameterize enzyme-constrained genome scale metabolic models (ecGEMs) facilitated by a Bayesian approach, which outperformed the default ecGEMs in predicting phenotypes and proteomes and enabled to explain phenotype differences among yeast species. The deep learning kcat prediction approach and automatic ecGEM construction pipeline would thus be a valuable tool to uncover the global trend of enzyme kinetics and physiological diversity, and to further elucidate cell metabolism on a large scale.

systems biology↗

bioRxiv2021

Predicting the Conformational Variability of Oncogenic GTP-bound G12D Mutated KRas-4B Proteins at Cell Membranes

KRas proteins are the largest family of mutated Ras isoforms, participating in a wide variety of cancers. Due to their importance, large effort is being carried out on drug development by small-molecule inhibitors. However, understanding protein conformational variability remains a challenge in drug discovery. In the case of the Ras family, their multiple conformational states can affect the binding of potential drug inhibitors. To overcome this challenge, we propose a computational framework based on combined all-atom Molecular Dynamics and Metadynamics simulations able to accurately access conformational variants of the target protein. We tested the methodology using a G12D mutated GTP bound oncogenic KRas-4B protein located at the interface of a DOPC/DOPS/cholesterol model anionic cell membrane. Two main orientations of KRas-4B at the anionic membrane have been obtained and explored. The corresponding angles have been taken as reliable reaction coordinates so that free-energy landscapes have been obtained by well-tempered metadynamics simulations, revealing the local and global minima of KRas-4B binding to the cell membrane, unvealing reactive paths of the system between the two preferential orientations and highlighting opportunities for targeting the unique metastable states through the identification of druggable pockets.

biophysics↗

bioRxiv2021

Lentiviral vector optimization enhances the expression and cytotoxicity of chimeric antigen receptors

Adoptive chimeric antigen receptor (CAR)-modified T or NK cells (CAR-T/NK) have emerged as a novel form of disease treatment. Lentiviral vectors (LVs) are commonly employed to engineer T/NK cells for the efficient expression of CARs. This study reported for the first time the influence of single-promoter and dual-promoter LVs on the CAR expression and cytotoxicity of engineered NK cells. Our results demonstrated that the selected CAR exhibits both a higher expression level and a higher coexpression concordance with the GFP reporter in HEK-293T or NK92 cells by utilizing the optimized single-promoter pCDHsp rather than the original dual-promoter pCDHdp. After puromycin selection, the pCDHsp produces robust CAR expression and enhanced in vitro cytotoxicity of engineered NK cells. Therefore, infection with a single-promoter pCDHsp lentivector is recommended to prepare CAR-engineered cells. This research will help to optimize the production of CAR-NK cells and improve their functional activity, to provide CAR-NK cell products with better and more uniform quality.

cell biology↗

bioRxiv2021

APRILE: Exploring the Molecular Mechanisms of Drug Side Effects with Explainable Graph Neural Networks

The majority of people over the age of 65 take two or more medications. While many individual drug side effects are known, polypharmacy side effects due to novel drug combinations poses great risk. Here, we present APRILE: an explainable artificial intelligence (XAI) framework that uses graph neural networks to explore the molecular mechanisms underlying polypharmacy side effects. Given a list of side effects and the pairs of drugs causing them, APRILE identifies a set of proteins (drug targets or non-targets) and associated Gene Ontology (GO) terms as mechanistic explanations of associated side effects. Using APRILE, we generate such explanations for 843,318 (learned) and 93,966 (novel) side effect-drug pair events, spanning 861 side effects (472 diseases, 485 symptoms and 9 mental disorders) and 20 disease cate-gories. We show that our two new metrics--pharmacogenomic information utilization and protein-protein interaction information utilization--provide quantitative estimates of mechanism complexity. Explanations were significantly consistent with state of the art disease-gene associations for 232/239 (97%) side effects. Further, APRILE generated new insights into molecular mechanisms of four diverse categories of adverse drug reactions: infection, metabolic diseases, gastrointestinal diseases, and mental disorders, including paradoxical side effects. We demonstrate the viability of discovering polypharmacy side effect mechanisms by training an XAI framework on massive biomedical data. Consequently, it facilitates wider and more reliable use of AI in healthcare.

bioinformatics↗

bioRxiv2021

Hexapeptide induces M2 macrophage polarization via the JAK1/STAT6 pathway to promote angiogenesis in bone repair

Angiogenesis is essential for successful bone defect repair. In normal tissue repair, the physiological inflammatory response is the main regulator of angiogenesis through the activity of macrophages and the cytokines secreted by them. In particular, M2 macrophages which secrete high levels of PDGF-BB are typically considered to promote angiogenesis. A hexapeptide [WKYMVm, (Trp-Lys-Tyr-Met-Val-D-Met-NH2)] has been reported to modulate inflammatory activities. However, the underlying mechanisms by which WKYMVm regulates macrophages remain unclear. In this study, the possible involvement by which WKYMVm induces the polarization of macrophages and affects their behaviors was evaluated. In vitro results showed that macrophages were induced to an M2 rather than M1 phenotype and the M2 phenotype was enhanced by WKYMVm through activation of the JAK1/STAT6 signaling pathway. It was also found that WKYMVm played an important role in the PDGF-BB production increase and proangiogenic abilities in M2 macrophages. Consistent with the results in vitro, the elevated M2/M0 ratio induced by WKYMVm enhanced the formation of new blood vessels in a femoral defect mouse model. In summary, these findings suggest that WKYMVm could be a promising alternative strategy for angiogenesis in bone repair by inducing M2 macrophage polarization.

biochemistry↗

bioRxiv2021

Accurate Virus Identification with Interpretable Raman Signatures by Machine Learning

Rapid identification of newly emerging or circulating viruses is an important first step toward managing the public health response to potential outbreaks. A portable virus capture device coupled with label-free Raman Spectroscopy holds the promise of fast detection by rapidly obtaining the Raman signature of a virus followed by a machine learning approach applied to recognize the virus based on its Raman spectrum. In this paper, we present a machine learning analysis on Raman spectra of human and avian viruses. A Convolutional Neural Network (CNN) classifier specifically designed for spectral data achieves very high accuracy for a variety of virus type or subtype identification tasks. In particular, it achieves 99% accuracy for classifying influenza virus type A vs. type B, 96% accuracy for classifying four subtypes of influenza A, 95% accuracy for differentiating enveloped and non-enveloped viruses, and 99% for differentiating avian coronavirus (infectious bronchitis virus, IBV) from other avian viruses. Furthermore, interpretation of neural net responses in the trained CNN model using a full-gradient algorithm highlights Raman spectral ranges that are most important to virus identification. By correlating ML-selected salient Raman ranges with the signature ranges of known biomolecules and chemical functional groups (e.g. amide, amino acid, carboxylic acid) we verify that our ML model effectively recognizes the Raman signatures of proteins, lipids and other vital functional groups present in different viruses and uses a weighted combination of these signatures to identify viruses. The accurate and interpretable machine learning model developed for Raman virus identification presents promising potential in a real-time virus detection system. Significance StatementA portable micro-fluidic platform for virus capture promises rapid enrichment and label-free optical identification of viruses by Raman spectroscopy. A large Raman dataset collected on a variety of viruses enables the training of machine learning (ML) models capable of highly accurate and sensitive virus identification. The trained ML models can then be integrated with the portable device to provide real-time virus detection and identification capability. We validate this conceptual framework by presenting highly accurate virus type and subtype identification results using a convolutional neural network to classify Raman spectra of viruses.

biophysics↗

bioRxiv2021

Differential Bcd activation of two hunchback promoters emerges from unified kinetics of enhancer-promoter interaction

Many eukaryotic genes contain alternative promoters with distinct expression patterns. How these promoters are differentially regulated remains elusive. Here, we apply single-molecule imaging to quantify the transcriptional regulation of two alternative promoters (P1 and P2) of the Bicoid (Bcd) target gene hunchback in syncytial blastoderm Drosophila embryos. Contrary to the previous notion that Bcd only activates P2, we find that Bcd activates both promoters via the same two enhancers. P1 activation is less frequent and requires binding of more Bcd molecules than P2 activation. Using a theoretical model to relate promoter activity to enhancer states, we show that the two promoters follow common transcription kinetics driven by sequential Bcd binding at the two enhancers. Bcd binding at either enhancer primarily activates P2, while P1 activation relies more on Bcd binding at both enhancers. These results provide a quantitative framework for understanding the kinetic mechanisms of complex eukaryotic gene regulation.

biophysics↗

bioRxiv2021

Biofilm self-patterning: mechanical force landscaping drives a collective reorientation cascade

In growing active matter systems, a large collection of engineered or living autonomous units metabolize free energy and create order at different length scales as they proliferate and migrate collectively. One such example is bacterial biofilms, which are surface-attached aggregates of bacterial cells embedded in an extracellular matrix. However, how bacterial growth coordinates with cell-surface interactions to create distinctive, long-range order in biofilms remains elusive. Here we report a collective cell reorientation cascade in growing Vibrio cholerae biofilms, leading to a differentially ordered, spatiotemporally coupled core-rim structure reminiscent of a blooming aster. Cell verticalization in the core generates differential growth that drives radial alignment of the cells in the rim, while the radially aligned rim in turn generates compressive stresses that expand the verticalized core. Such self-patterning disappears in adhesion-less mutants but can be restored through opto-manipulation of growth. Agent-based simulations and two-phase active nematic modeling reveal the strong interdependence of the driving forces for the differential ordering. Our findings provide insight into the collective cell patterning in bacterial communities and engineering of phenotypes and functions of living active matter.

biophysics↗

bioRxiv2021

Metagenome-genome-wide association studies reveal human genetic impact on the oral microbiome

The oral microbiota contains billions of microbial cells, which could contribute to diseases in a number of body sites. Challenged by eating, drinking and dental hygiene on a daily basis, the oral microbiota is regarded as highly dynamic. Here, we report significant human genomic associations with the oral metagenome from more than 1,915 individuals, for both the tongue dorsum and saliva. We identified five genetic loci associated with oral microbiota at study-wide significance (p < 3.16 x 10-11). Four of the five associations were well replicated in an independent cohort of 1,439 individuals: rs1196764 at APPL2 with Prevotella jejuni, Oribacterium uSGB 3339 and Solobacterium uSGB 315; rs3775944 at the serum uric acid transporter SLC2A9 with Oribacterium uSGB 1215, Oribacterium uSGB 489 and Lachnoanaerobaculum umeaense; rs4911713 near OR11H1 with species F0422 uSGB 392; and rs36186689 at LOC105371703 with Eggerthia. Further analyses confirmed 84% (386/455 for tongue dorsum) and 85% (391/466 for saliva) of genetic-microbiota associations including 6 genome-wide significant associations mutually validated between the two niches. Human genetics accounted for at least 10% of oral microbiome compositions between individuals. Machine learning models showed that polygenetic risk score dominated over oral microbiome in predicting predisposing risk of dental diseases such as dental calculus and gingival bleeding. These findings indicate that human genetic differences are one explanation for a stable or recurrent oral microbiome in each individual.

genetics↗