Biology subjects

Fu, W.

Publications and source records attributed to Fu, W..

At least 73 records · Page 4Linked to original sources

bioRxiv2020

Embedding covariate adjustments in tree-based automated machine learning for biomedical big data analyses

BackgroundA typical task in bioinformatics consists of identifying which features are associated with a target outcome of interest and building a predictive model. Automated machine learning (AutoML) systems such as the Tree-based Pipeline Optimization Tool (TPOT) constitute an appealing approach to this end. However, in biomedical data, there are often baseline characteristics of the subjects in a study or batch effects that need to be adjusted for in order to better isolate the effects of the features of interest on the target. Thus, the ability to perform covariate adjustments becomes particularly important for applications of AutoML to biomedical big data analysis. ResultsWe present an approach to adjust for covariates affecting features and/or target in TPOT. Our approach is based on regressing out the covariates in a manner that avoids leakage during the cross-validation training procedure. We then describe applications of this approach to toxicogenomics and schizophrenia gene expression data sets. The TPOT extensions discussed in this work are available at https://github.com/EpistasisLab/tpot/tree/v0.11.1-resAdj. ConclusionsIn this work, we address an important need in the context of AutoML, which is particularly crucial for applications to bioinformatics and medical informatics, namely covariate adjustments. To this end we present a substantial extension of TPOT, a genetic programming based AutoML approach. We show the utility of this extension by applications to large toxicogenomics and differential gene expression data. The method is generally applicable in many other scenarios from the biomedical field.

bioinformatics↗

bioRxiv2020

CAR Macrophages for SARS-CoV-2 Immunotherapy

Targeted therapeutics for the treatment of coronavirus disease 2019 (COVID-19), especially severe cases, are currently lacking. As macrophages have unique effector functions as a first-line defense against invading pathogens, we genetically armed human macrophages with chimeric antigen receptors (CARs) to reprogram their phagocytic activity against SARS-CoV-2. After investigation of CAR constructs with different intracellular receptor domains, we found that although cytosolic domains from MERTK (CARMERTK) did not trigger antigen-specific cellular phagocytosis or killing effects, unlike those from MEGF10, FcR{gamma} and CD3{zeta} did, these CARs all mediated similar SARS-CoV-2 clearance in vitro. Notably, we showed that CARMERTK macrophages reduced the virion load without upregulation of proinflammatory cytokine expression. These results suggest that CARMERTK drives an immunologically silent scavenger effect in macrophages and pave the way for further investigation of CARs for the treatment of individuals with COVID-19, particularly those with severe cases at a high risk of hyperinflammation.

bioengineering↗

bioRxiv2020

A novel biparatopic antibody-ACE2 fusion that blocks SARS-CoV-2 infection: implications for therapy

In the absence of a proven effective vaccine preventing infection by SARS-CoV-2, or a proven drug to treat COVID-19, the positive results of passive immune therapy using convalescent serum provides a strong lead. We have developed a new class of tetravalent, biparatopic therapy, 89C8-ACE2. It combines the specificity of a monoclonal antibody (89C8) that recognizes the relatively conserved N-terminal domain (NTD) of the viral S glycoprotein, and the ectodomain of ACE2, which binds to the receptor-binding domain (RBD) of S. This molecule shows exceptional performance in vitro, inhibiting the interaction of recombinant S1 to ACE2 and transduction of ACE2-overexpressing cells by S-pseudotyped lentivirus with IC50s substantially below 100 pM, and with potency approximately 100-fold greater than ACE2-Fc itself. Moreover, 89C8-ACE2 was able to neutralize authentic virus infection in a standard assay at low nanomolar concentrations, making this class of molecule a promising lead for therapeutic applications.

immunology↗

bioRxiv2020

Influence of isolation by environment and landscape heterogeneity on genetic structure of wild rice Zizania latifolia along a latitudinal gradient

Global aquatic habitats are undergoing rapid degradation and fragmentation as a result of land-use change and climate change. Understanding the genetic variability and adaptive potential of aquatic plant species is thus important for conservation purposes. In this study, we investigated the role of environment, landscape heterogeneity and geographical distance in shaping the genetic structure of 28 natural populations of Zizania latifolia (Griseb.) Turcz. Ex Stapf in China based on 25 microsatellite markers. Genetic structure was investigated by analysis of molecular variance (AMOVA), estimation of FST, Bayesian clustering and Thermodynamic Integration (TI) methods. Isolation by environment (IBE), isolation by resistance (IBR) and isolation by distance (IBD) hypotheses were compared using a reciprocal causal model (RCM). Further, generalized linear models and spatially explicit mixed models, by using geographic, landscape and genetic variables, were developed to elucidate the role of environment in driving Z. latifolia genetic diversity. The genetic differentiation across all populations was high: FST = 0.579; Opt = 0.578. RCM exclusively supported IBE in shaping genetic structuring, only partial support for IBR, but not for IBD. Maximum temperature of the warmest month and precipitation seasonality were the plausible parameters responsible for genetic diversity. After controlling for spatial effect and landscape complexity, precipitation seasonality was significantly associated with genetic diversity. Based on these findings, genetic structure of Z. latifolia across China seem to be as a result of local adaptation. Environmental gradient and topographical barriers, rather than geographical isolation, influence genetic differentiation of aquatic species across China resulting in instances of local adaptation.

ecology↗

bioRxiv2020

Tracing the origin of a new organ by inferring the genetic basis of rumen evolution

The rumen is the hallmark organ of ruminants and hosts a diverse ecosystem of microorganisms that facilitates efficient digestion of plant fibers. We used 897 transcriptomes from three Cetartiodactyla lineages: ruminants, camels and cetaceans, as well as data from ruminant comparative genomics and functional assays to explore the genetic basis of rumen origin and evolution. Comparative analyses reveal that the rumen and the first-chamber stomachs of camels and cetaceans shared a common tissue origin from the esophagus. The rumen recruited genes from other tissues/organs and up-regulated many esophagus genes to aquire functional innovations involving epithelium absorption, improvement of the ketone body metabolism and regulation of microbial community. These innovations involve such genetic changes as ruminant-specific conserved elements, newly evolved genes and positively selected genes. Our in vitro experiements validate the functions of one enhancer, one positively selected gene and two newly evolved antibacterial genes. Our study provides novel insights into the origin and evolution of a complex organ.

evolutionary biology↗

bioRxiv2020

Potent neutralization of 2019 novel coronavirus by recombinant ACE2-Ig

2019-nCoV, which is a novel coronavirus emerged in Wuhan, China, at the end of 2019, has caused at least infected 11,844 as of Feb 1, 2020. However, there is no specific antiviral treatment or vaccine currently. Very recently report had suggested that novel CoV would use the same cell entry receptor, ACE2, as the SARS-CoV. In this report, we generated a novel recombinant protein by connecting the extracellular domain of human ACE2 to the Fc region of the human immunoglobulin IgG1. An ACE2 mutant with low catalytic activity was also used in the study. The fusion proteins were then characterized. Both fusion proteins has high affinity binding to the receptor-binding domain (RBD) of SARS-CoV and 2019-nCoV and exerted desired pharmacological properties. Moreover, fusion proteins potently neutralized SARS-CoV and 2019-nCoV in vitro. As these fusion proteins exhibit cross-reactivity against coronaviruses, they could have potential applications for diagnosis, prophylaxis, and treatment of 2019-nCoV.

bioengineering↗

bioRxiv2020

RNA-binding protein Syncrip regulates Starvation-Induced Hyperactivity in adult Drosophila

How to respond to starvation determines fitness. One prominent behavioral response is increased locomotor activities upon starvation, also known as Starvation-Induced Hyperactivity (SIH). SIH is paradoxical as it promotes food seeking but also increases energy expenditure. Despite its importance in fitness, the genetic contributions to SIH as a behavioral trait remains unexplored. Here, we examined SIH in the Drosophila melanogaster Genetic Reference Panel (DGRP) and performed genome-wide association studies. We identified 23 significant loci, corresponding to 14 genes, significantly associated with SIH in adult Drosophila. Gene enrichment analyses indicated that genes encoding ion channels and mRNA binding proteins (RBPs) were most enriched in SIH. We are especially interested in RBPs because they provide a potential mechanism to quickly change protein expression in response to environmental challenges. Using RNA interference, we validated the role of syp in regulating SIH. syp encodes Syncrip (Syp), an RBP. While ubiquitous knockdown of syp led to semi-lethality in adult flies, adult flies with neuron-specific syp knockdown were viable and exhibited decreased SIH. Using the Temporal and Regional Gene Expression Targeting (TARGET) system, we further confirmed the role of Syp in adult neurons in regulating SIH. To determine how syp is regulated by starvation, we performed RNA-seq using the heads of flies maintained under either food or starvation conditions. RNA-seq analyses revealed that syp was alternatively spliced under starvation while its expression level was unchanged. We further generated an alternatively-spliced-exon-specific knockout (KO) line and found that KO flies showed reduced SIH. Together, this study demonstrates a significant genetic contribution to SIH as a behavioral trait, identifies syp as a SIH gene, and highlights the significance of RBPs and post-transcriptional processes in the brain in regulating behavioral responses to starvation. Author summaryAnimals living in the wild often face periods of starvation. How to physiologically and behaviorally respond to starvation is essential for survival. One behavioral response is Starvation-Induced Hyperactivity (SIH). We used the Drosophila melanogaster Genetic Reference Panel, derived from a wild population, to study the genetic basis of SIH. Our results show that there is a significant genetic contribution to SIH in this population, and that genes encoding RNA binding proteins (RBPs) are especially important. Using RNA interference and the TARGET system, we confirmed the role of an RBP Syp in adult neurons in SIH. Using RNA-seq and Western blotting, we found that syp was alternatively spliced under starvation while its expression level was unchanged. Further studies from syp exon-specific knockout flies showed that alternative splicing involving two exons in syp was important for SIH. Together, this study identifies syp as a SIH gene and highlights an essential role of post-transcriptional modification in regulating this behavior.

genomics↗

bioRxiv2019

Antibody Bank: the Synthetic Cell Repertoire for Immunotherapy

The immune surveillance hypothesis is regarded as the intellectual underpinning of cancer immunology. We find that engineered T cell library displaying CAR constructs, where the antibody repertoires took the place of the antigen-recognition region, has the capacity to recognize nonself-antigens and showed antigen dependent clonal expanding and persistence. This process coupled with long-lasting anti-tumour effect in a broad spectrum of epithelial tumours. Moreover, CAR-T library showed robust immunologic memory and maintenance of diversity to recognize mutated or evolved tumours in both immunodeficiency and immunoresponsive murine tumor models. Large CAR library was further developed in logic-gated NK-92 cells to develop an off-the shelf therapeutic library. Thus, the biological behaviour of synthetic cell library is highly accordance with the immune surveillance function, which proved the way to rebuilt functional MHC unrestricted cellular immunity in vitro using synthetic method.

bioengineering↗

bioRxiv2019

AutoGenome: An AutoML Tool for Genomic Research

Deep learning have made great successes in traditional fields like computer vision (CV), natural language processing (NLP) and speech processing. Those achievements greatly inspire researchers in genomic study and make deep learning in genomics a very hot topic. Convolutional neural network (CNN) and recurrent neural network (RNN) are frequently used for genomic sequence prediction problems; multiple layer perception (MLP) and auto-encoders (AE) are frequently used for genomic profiling data like RNA expression data and gene mutation data. Here, we introduce a new neural network architecture, named residual fully-connected neural network (RFCN) and demonstrate its advantage for modeling genomic profiling data. We further incorporate AutoML algorithms and implement AutoGenome, an end-to-end automated genomic deep learning framework. By utilizing the proposed RFCN architectures, automatic hyper-parameter search and neural architecture search algorithms, AutoGenome can train high-performance deep learning models for various kinds of genomic profiling data automatically. To make researchers better understand the trained models, AutoGenome can assess the feature importance and export the most important features for supervised learning tasks, and the representative latent vectors for unsupervised learning tasks. We envision AutoGenome to become a popular tool in genomic studies.

genomics↗

bioRxiv2019

TIGIT-Fc Promote Immune Tolerance at the Feto-maternal Interface

The perfect synchronization of maternal immune-endocrine mechanisms and those of the foetus is necessary for a successful pregnancy. In this report, decidual immune cells at the maternal-foetal interface were detected that expressed TIGIT (T cell immunoreceptor with Ig and ITIM domains), which is a co-inhibitory receptor that triggers immunological tolerance. We generated recombinant TIGIT-Fc fusion proteins by linking the extracellular domain of TIGIT and silent Fc fragments. The treatment with TIGIT-Fc of human decidual dendritic cells (dDCs) increased the production of interleukin 10 and induced the dDCs to powerfully polarize the decidual CD4+ T cells towards a classic TH2 phenotype. The administration of TIGIT-Fc to CBA/J pregnant mice at preimplantation induced CD4+ forkhead box P3+ (Foxp3+) regulatory T cells and tolerogenic dendritic cells and increased pregnancy rates in a mouse model of abortive stress. Moreover, we proposed that progesterone play a direct role in the transcriptional regulation of the TIGIT gene in decidual immune cell subsets. The results suggested the therapeutic potential of the TIGIT-Fc fusion protein in reinstating immune tolerance in failing pregnancies.

immunology↗

bioRxiv2019

The Bovine Genome Variation Database (BGVD): Integrated Web-database for Bovine Sequencing Variations and Selective Signatures

Next-generation sequencing has yielded a vast amount of cattle genomic data for the global characterization of population genetic diversity and the identification of regions of the genome under natural and artificial selection. However, efficient storage, querying and visualization of such large datasets remain challenging. Here, we developed a comprehensive Bovine Genome Variation Database (BGVD, http://animal.nwsuaf.edu.cn/BosVar) that provides six main functionalities: Gene Search, Variation Search, Genomic Signature Search, Genome Browser, Alignment Search Tools and the Genome Coordinate Conversion Tool. The BGVD contains information on genomic variations comprising [~]60.44 M SNPs, [~]6.86 M indels, 76,634 CNV regions and signatures of selective sweeps in 432 samples from modern cattle worldwide. Users can quickly retrieve distribution patterns of these variations for 54 cattle breeds through an interactive source of breed origin map using a given gene symbol or genomic region for any of the three versions of the bovine reference genomes (ARS-UCD1.2, UMD3.1.1, and Btau 5.0.1). Signals of selection are displayed as Manhattan plots and Genome Browser tracks. To further investigate and visualize the relationships between variants and signatures of selection, the Genome Browser integrates all variations, selection data and resources from NCBI, the UCSC Genome Browser and AnimalQTLdb. Collectively, all these features make the BGVD a useful archive for in-depth data mining and analyses of cattle biology and cattle breeding on a global scale.

genomics↗

bioRxiv2019

Enhancing single-cell cellular state inference by incorporating molecular network features

In biological systems, genes function in conjunction rather than in isolation. However, traditional single-cell RNA-seq (scRNA-seq) analyses heavily rely on the transcriptional similarity of individual genes, ignoring the inherent gene-gene interactions. Here, we present SCORE, a network-based method, which incorporates the validated molecular network features to infer cellular states. Using real scRNA-seq datasets, SCORE outperforms existing methods in accuracy, robustness, scalability, data integration and removal of batch effect. When applying SCORE to a newly generated human ileal scRNA-seq dataset, we identified several novel stem/progenitor clusters, including a Cripto-1+ cluster. Moreover, two distinct groups of goblet cells were identified and only one of them tended to secrete mucus. Besides, we found that the recently identified BEST4+OTOP2+ microfold cells also highly expressed CFTR, which is different from their colonic counterparts. In summary, SCORE enhances cellular state inference by simulating the dynamic changes of molecular networks, providing more biological insights beyond statistical interpretations.

bioinformatics↗

bioRxiv2019

Treatment of Murine Lupus with TIGIT-Ig

The TIGIT (T cell immunoreceptor with Ig and ITIM domains) protein is a co-inhibitory receptor that has been reported to suppress autoreactive T and B cells to trigger immunological tolerance. We generated a new recombinant protein by connecting the extracellular domain of murine TIGIT to the Fc region of the rat immunoglobulin IgG2a. The fusion protein was then characterized. The results suggested that among mice with lupus that were treated with the TIGIT-Ig fusion protein, the onset of proteinuria was delayed, serum concentrations of autoantibodies, such as antinuclear antibodies, were reduced without a decrease in the total IgG concentrations, and the survival rate was significantly increased compared to those of the controls. In conclusion, TIGIT-Ig administration showed promising results for both the prevention and treatment of autoimmune diseases in mice. This indicates that treatment with recombinant human TIGIT-Ig shows promise as an effective way to treat human autoimmune diseases.

immunology↗

bioRxiv2019

Fexofenadine inhibits TNF signaling through targeting to cytosolic phospholipase A2 and is therapeutic against autoimmune diseases

TNF- signaling plays a central role in the pathogenesis of various diseases, particularly autoimmune diseases. Screening of a library composed of FDA approved drugs led to the identification of Terfenadine and its active metabolite Fexofenadine as inhibitors of TNF- signaling. Both Fexofenadine and Terfenadine inhibited TNF/NF-{kappa}B signaling in vitro and in vivo, and ameliorated disease symptoms in various autoimmune disease models, including TNF- transgenic mice, collagen-induced arthritis, and inflammatory bowel disease. Subsequent studies identified cytosolic phospholipase A2 (cPLA2) as a novel target of Fexofenadine. Fexofenadine blocked TNF-stimulated cPLA2 activity and arachidonic acid production through binding to catalytic domain 2 of cPLA2 and inhibition of its phosphorylation on Ser-505. Further, deletion of cPLA2 abolished Fexofenadines anti-TNF activity. Collectively, these findings not only provide new insights into the understanding of Fexofenadine action and underlying mechanism, but also provide new therapeutic interventions for various TNF- and cPLA2-associated pathologies and conditions, particularly autoimmune diseases.

molecular biology↗

bioRxiv2018

Scaling tree-based automated machine learning to biomedical big data with a dataset selector

Automated machine learning (AutoML) systems are helpful data science assistants designed to scan data for novel features, select appropriate supervised learning models and optimize their parameters. For this purpose, Tree-based Pipeline Optimization Tool (TPOT) was developed using strongly typed genetic programming to recommend an optimized analysis pipeline for the data scientists prediction problem. However, TPOT may reach computational resource limits when working on big data such as whole-genome expression data. We introduce two new features implemented in TPOT that helps increase the systems scalability: Dataset selector and Template. Dataset selector (DS) provides the option to specify subsets of the features as separate datasets, assuming the signals come from one or more of these specific data subsets. Built in at the beginning of each pipeline structure, DS reduces the computational expense of TPOT to only evaluate on a smaller subset of data rather than the entire dataset. Consequently, DS increases TPOTs efficiency in application on big data by slicing the dataset into smaller sets of features and allowing genetic programming to select the best subset in the final pipeline. Template enforces type constraints with strongly typed genetic programming and enables the incorporation of DS at the beginning of each pipeline. We show that DS and Template help reduce TPOT computation time and may provide more interpretable results. Our simulations show TPOT-DS significantly outperforms a tuned XGBoost model and standard TPOT implementation. We apply TPOT-DS to real RNA-Seq data from a study of major depressive disorder. Independent of the previous study that identified significant association with depression severity of the enrichment scores of two modules, in an automated fashion, TPOT-DS corroborates that one of the modules is largely predictive of the clinical diagnosis of each individual.\n\nAuthor SummaryBig data have recently become prevalent in many fields including meteorology, complex physics simulations, large scale imaging, genomics, biomedical research, environmental research and more. TPOT is a Python Automated Machine Learning (AutoML) tool that uses genetic programming to optimize machine learning pipelines for analyzing biomedical data. However, like other AutoML tools, when analyzing big data, the early implementations of TPOT face the challenges of long runtime, high computational expense as well complex pipeline with low interpretability. Here, we develop two novel features for TPOT, Dataset Selector and Template, that leverage domain knowledge, greatly reduce the computational expense and flexibly extend TPOTs application to biomedical big data analysis.

genomics↗

bioRxiv2018

Generating a sequence map of the pig pan-genome

Pigs (Sus scrofa) exhibit diverse phenotypes in different breeds shaped by the combined effects of various local adaptation and artificial selection. To comprehensively characterize the genetic diversity of pigs, we construct a pig pan-genome by comparing genome assemblies of 11 representative pig breeds with the reference genome (Sscrofa11.1). Approximately 72.5 Mb non-redundant sequences were identified as pan-sequences which were absent from the Sscrofa11.1. On average, 41.7 kb of spurious heterozygous SNPs per individual are removed and 12.9 kb novel SNPs per individual are recovered using pan-genome as the reference for SNP calling, thereby providing enhanced resolution for genetic diversity in pigs. Homolog annotation and analysis using RNA-seq and Hi-C data indicate that these pan-sequences contain protein-coding regions and regulatory elements. These pan-sequences can further improve the interpretation of local 3D structure. The pan-genome as well as the accompanied web-based database will serve as a primary resource for exploration of genetic diversity and promote pig breeding and biomedical research.

genomics↗

bioRxiv2018

Targeting RyR2 with a phosphorylation site-specific nanobody Reverses Dysfunction of Failing Cardiomyocytes in Rat

Chronic PKA phosphorylation of RyR2 has been shown to increased diastolic SR Ca2+ leak and lead to cardiac dysfunction. Since the change of phosphorylation level of RyR2 is a biomarker of failing heart, we attempted to verify the hypothesis that intracellular gene delivery of a RyR2 targeting phosphorylation site-specific nanobody could preserve contractility of failing myocardium. In present study, we acquired the RyR2-specific nanobodies from a phage display library which are variable domains of camellidae heavy chain-only antibodies (VHH). One of the monoclonal nanobodies, AR185, inhibiting RyR2 phosphorylation in an in vitro assay was then chosen for further investigation. We investigated the potential of adeno-associated virus (AAV)-9-mediated cardiac expression of AR185 against post-ischemic heart failure. Adeno-associated virus gene delivery elevated the intracellular expression AR185 protein in the ischemic heart failure model of rats, and this treatment normalized the systolic and diastolic dysfunction of the failing myocardium in vivo and in vitro by reversing myocardial Ca2+ handling. Furthermore, AR185 gene transfer to failing cardiomyocytes reduced the frequency of sarcoplasmic reticulum (SR) calcium leak, thereby restoring the attenuated intracellular calcium transients and SR calcium load. Moreover, AR185 gene transfer inhibited PKA phosphorylation of RyR2 in failing cardiomyocytes. Our results provided strong pre-clinical experimental evidence of the cardiac expression of RyR2 nanobody with AAV9 vectors as a promising therapeutic strategy for ischemic heart failure.

pathology↗

bioRxiv2018

A familial case of Syndactyly type IV due to a novel duplication of ~222.23 kb covering exons 2-17 of the LMBR1 gene: a case report

Syndactyly is one of the most frequent hereditary limb malformations with clinical and genetical complexity. Autosomal dominant Syndactyly type IV (SD4) is a very rare form of syndactyly, occurring as a result of heterozygous mutation in an SHH regulatory element (ZRS) that resides in intron 5 of the LMBR1 gene on chromosome 7q36.3. The SD4 is characterized by complete cutaneous syndactyly of all fingers, cup-shaped hands due to flexion of the fingers and accompanied by polydactyly. Here, we firstly reported a big Chinese family, manifesting cup-shaped hands consistent with SD4 and intrafamilial heterogeneity in clinical phenotype of tibial and fibulal shortening, triphalangeal thumb-polysyndactyly syndrome (TPTPS). Genetically, we identified a novel duplication of [~]222.23 kb covering exons 2-17 of the LMBR1 gene in this family by next generation sequencing. This case expands our new clinical understanding of SD4 phenotype.

genetics↗