Biology subjects

Wu, Y.

Publications and source records attributed to Wu, Y..

At least 1,081 records · Page 60Linked to original sources

bioRxiv2018

Forecasting Future Humphrey Visual Fields Using Deep Learning

PurposeTo determine if deep learning networks could be trained to forecast a future 24-2 Humphrey Visual Field (HVF).\n\nDesignRetrospective database study.\n\nParticipantsAll patients who obtained a HVF 24-2 at the University of Washington.\n\nMethodsAll datapoints from consecutive 24-2 HVFs from 1998 to 2018 were extracted from a University of Washington database. Ten-fold cross validation with a held out test set was used to develop the three main phases of model development: model architecture selection, dataset combination selection, and time-interval model training with transfer learning, to train a deep learning artificial neural network capable of generating a point-wise visual field prediction.\n\nMain outcome measuresMean absolute error (MAE) and difference in Mean Deviation (MD) between predicted and actual future HVF.\n\nResultsMore than 1.7 million perimetry points were extracted to the hundredth decibel from 32,443 24-2 HVFs. The best performing model with 20 million trainable parameters, CascadeNet-5, was selected. The overall MAE for the test set was 2.47 dB (95% CI: 2.45 dB to 2.48 dB). The 100 fully trained models were able to successfully predict progressive field loss in glaucomatous eyes up to 5.5 years in the future with a correlation of 0.92 between the MD of predicted and actual future HVF (p < 2.2 = 10-16) and an average difference of 0.41 dB.\n\nConclusionsUsing unfiltered real-world datasets, deep learning networks show an impressive ability to not only learn spatio-temporal HVF changes but also to generate predictions for future HVFs up to 5.5 years, given only a single HVF.

bioinformatics↗

bioRxiv2018

A unifying framework for summary statistic imputation

Imputation has been widely utilized to aid and interpret the results of Genome-Wide Association Studies(GWAS). Imputation can increase the power to identify associations when the causal variant was not directly observed or typed in the GWAS. There are two broad classes of methods for imputation. The first class imputes the genotypes at the untyped variants given the genotypes at the typed variants and then performs a statistical test of association at the imputed variants. The second class of methods, summary statistic imputation, directly imputes the association statics at the untyped variants given the association statistics observed at the typed variants. This second class of methods is appealing as it tends to be computationally efficient while only requiring the summary statistics from a study while the former class requires access to individual-level data that can be difficult to obtain. The statistical properties of these two classes of imputation methods have not been fully understood. In this paper, we show that the two classes of imputation methods are equivalent, i.e., have identical asymptotic multivariate normal distributions with zero mean and minor variations in the covariance matrix, under some reasonable assumptions. Using this equivalence, we can understand the effect of imputation methods on power. We show that a commonly employed modification of summary statistic imputation that we term summary statistic imputation with variance re-weighting generally leads to a loss in power. On the other hand, our proposed method, summary statistic imputation without performing variance re-weighting, fully accounts for imputation uncertainty while achieving better power.

bioinformatics↗

bioRxiv2018

MRI2MRI: A deep convolutional network that accurately transforms between brain MRI contrasts

Different brain MRI contrasts represent different tissue properties and are sensitive to different artifacts. The relationship between different contrasts is therefore complex and nonlinear. We developed a deep convolutional network that learns the mapping between different MRI contrasts. Using a publicly available dataset, we demonstrate that this algorithm accurately transforms between T1- and T2-weighted images, proton density images, time-of-flight angiograms, and diffusion MRI images. We demonstrate that these transformed images can be used to improve spatial registration between MR images of different contrasts.

neuroscience↗

bioRxiv2018

Transcriptome Landscape of Human Oocytes and Granulosa Cells Throughout Folliculogenesis

Folliculogenesis is a highly regulated process that involves bidirectional interactions of the oocytes and surrounding granulosa cells (GCs). Little is unknown, however, about the transcriptomic profiles of human oocytes and GCs throughout folliculogenesis. Here we performed a high resolution RNA-Seq of human oocytes and GCs at each follicular stage, which revealed unique transcriptional profiles, stage-specific signature genes, oocyte- and GC-derived genes that reflect ovarian reserve. We identified reciprocal cell-to-cell interactions between oocytes and GCs, including NOTCH, TGF-{beta} signaling and gap junctions and determined the expression patterns of maternal-effect genes involved in folliculogenesis and early embryogenesis. Finally, we demonstrated robust differences between human and mice oocyte transcriptomes. This is the first comprehensive overview of the transcriptomic signatures governing the stepwise human folliculogenesis in-vivo that provides a valuable resource for basic and translational research in human reproductive biology.

cell biology↗

bioRxiv2018

Novel susceptibility loci and genetic regulation mechanisms for type 2 diabetes

We conducted a meta-analysis of genome-wide association studies (GWAS) with [~]16 million genotyped/imputed genetic variants in 62,892 type 2 diabetes (T2D) cases and 596,424 controls of European ancestry. We identified 139 common and 4 rare (minor allele frequency < 0.01) variants associated with T2D, 42 of which (39 common and 3 rare variants) were independent of the known variants. Integration of the gene expression data from blood (n = 14,115 and 2,765) and other T2D-relevant tissues (n = up to 385) with the GWAS results identified 33 putative functional genes for T2D, three of which were targeted by approved drugs. A further integration of DNA methylation (n = 1,980) and epigenomic annotations data highlighted three putative T2D genes (CAMK1D, TP53INP1 and ATP5G1) with plausible regulatory mechanisms whereby a genetic variant exerts an effect on T2D through epigenetic regulation of gene expression. We further found evidence that the T2D-associated loci have been under purifying selection.

genetics↗

bioRxiv2018

CLADES: A Classification-based Machine Learning Method for Species Delimitation from Population Genetic Data

Species are considered to be the basic unit of ecological and evolutionary studies. Since multi-locus genomic data are becoming increasingly available, there has been considerable interests in the use of DNA sequence data to delimit species. In this paper, we show that machine learning can be used for species delimitation. There exists no species delimitation methods that are based on machine learning. Our method treats the species delimitation problem as a classification problem. It is a problem of identifying the category of a new observation on the basis of training data. Extensive simulation is first conducted over a broad range of evolutionary parameters for training purpose. Each pair of known populations are combined to form training samples with a label of \"same species\" or \"different species\". We use Support Vector Machine (SVM) to train a classifier using a set of summary statistics computed from training samples as features. The trained classifier can classify a test sample to two outcomes: \"same species\" or \"different species\". Given multi-locus genomic data of multiple related organisms or populations, our method (called CLADES) performs species delimitation by first classifying pairs of populations. CLADES then delimits species by maximizing the likelihood of species assignment for multiple populations. CLADES is evaluated through extensive simulation and also tested on real genetic data. We show that CLADES is both accurate and efficient for species delimitation when compared with existing methods. CLADES can be useful especially when existing methods have difficulty in delimitation, e.g. with short species divergence time and gene flow.

evolutionary biology↗

bioRxiv2018

Identifying gene targets for brain-related traits using transcriptomic and methylomic data from blood

Understanding the difference in genetic regulation of gene expression between brain and blood is important for discovering genes associated with brain-related traits and disorders. Here, we estimate the correlation of genetic effects at the top associated cis-expression (cis-eQTLs or cis-mQTLs) between brain and blood for genes expressed (or CpG sites methylated) in both tissues, while accounting for errors in their estimated effects (rb). Using publicly available data (n = 72 to l,366), we find that the genetic effects of cis-eQTLs (PeQTL < 5x10-8) or mQTLs (PmQTL < 1x10-10) are highly correlated between independent brain and blood samples ([Formula] with SE = 0.015 for cis-eQTL and [Formula] with SE = 0.006 for cis-mQTLs). Using meta-analyzed brain eQTL/mQTL data (n = 526 to 1,194), we identify 61 genes and 167 DNA methylation (DNAm) sites associated with 4 brain-related traits and disorders. Most of these associations are a subset of the discoveries (97 genes and 295 DNAm sites) using data from blood with larger sample sizes (n = l,980 to 14,115). We further find that cis-eQTLs with tissue-specific effects are approximately uniformly distributed across all the functional annotation categories, and that mean difference in gene expression level between brain and blood is almost independent of the difference in the corresponding cis-eQTL effect. Our results demonstrate the gain of power in gene discovery for brain-related phenotypes using blood cis-eQTL or cis-mQTL data with large sample sizes.

genetics↗

bioRxiv2018

A novel GATA-binding protein 4 gene variation associated with familial atrial septal defect

Atrial septal defect (ASD) is the most common congenital heart defect. Part of ASD exhibits familial predisposition, but the genetic mechanism remains largely unknown. In the current study, we use multiple methods to identify and confirm the gene associated with a familial ASD. Chromosomal microarray analyses, whole exome sequencing, Sanger sequencing, multiple bioinformatics programs, in silico protein structure modeling and molecular dynamics simulation were performed to predict the pathogenic of the variant gene. Dual-Luciferase reporter gene assay was performed to evaluate the influence of downstream target gene of the target variation. A novel, heterozygous, missense variant GATA-binding protein 4 (GATA4):c.958C>T, p.R320W was identified. An autosomal dominant inheritance pattern with incomplete penetrance was observed in the family. Multiple prediction indicate the variant in GATA4 to be deleterious. Molecular dynamics simulation further revealed that the variation of p.R320W could prevent the zinc finger of GATA4 from interacting with the DNA. Dual-Luciferase reporter assay demonstrated a significant decrease in transcriptional activity (0.90{+/-}0.099 vs 1.50{+/-}0.079, p = 0.001) of the variant GATA4 compared with the wild type. We believe the novel variation of GATA4 (c.958C>T, p.R320W) with a pattern of incomplete inheritance that may be highly associated with this familial ASD. The finding enriched our knowledge of variations that may associated with ASD.

genetics↗

bioRxiv2018

Visualizing single-cell RNA-seq datasets with Similarity Weighted Nonnegative Embedding (SWNE)

High throughput single-cell gene expression profiling has enabled the characterization of novel cell types and developmental trajectories. Visualizing these datasets is crucial to biological interpretation, and the most popular method is t-Stochastic Neighbor embedding (t-SNE), which visualizes local patterns better than other methods, but often distorts global structure, such as distances between clusters. We developed Similarity Weighted Nonnegative Embedding (SWNE), which enhances interpretation of datasets by embedding the genes and factors that separate cell states alongside the cells on the visualization, captures local structure better than t-SNE and existing methods, and maintains fidelity when visualizing global structure. SWNE uses nonnegative matrix factorization to decompose the gene expression matrix into biologically relevant factors, embeds the cells, genes and factors in a 2D visualization, and uses a similarity matrix to smooth the embeddings. We demonstrate SWNE on single cell RNA-seq data from hematopoietic progenitors and human brain cells.

bioinformatics↗

bioRxiv2018

Generating perfusion maps from structural optical coherence tomography with artificial intelligence

Despite advances in artificial intelligence (AI), its application in medical imaging has been burdened and limited by expert-generated labels. We used images from optical coherence tomography angiography (OCTA), a relatively new imaging modality that measures retinal blood flow, to train an AI algorithm to generate flow maps from standard optical coherence tomography (OCT) images, exceeding the ability and bypassing the need for expert labeling. Deep learning was able to infer flow from single structural OCT images with similar fidelity to OCTA and significantly better than expert clinicians (P < 0.00001). Our model allows generating flow maps from large volumes of previously collected OCT data in existing clinical trials and clinical practice. This finding demonstrates a novel application of AI to medical imaging, whereby subtle regularities between different modalities are used to image the same body part and AI is used to generate detailed inferences of tissue function from structure imaging.

neuroscience↗

bioRxiv2018

Genome-wide study identifies 611 loci associated with risk tolerance and risky behaviors

Humans vary substantially in their willingness to take risks. In a combined sample of over one million individuals, we conducted genome-wide association studies (GWAS) of general risk tolerance, adventurousness, and risky behaviors in the driving, drinking, smoking, and sexual domains. We identified 611 approximately independent genetic loci associated with at least one of our phenotypes, including 124 with general risk tolerance. We report evidence of substantial shared genetic influences across general risk tolerance and risky behaviors: 72 of the 124 general risk tolerance loci contain a lead SNP for at least one of our other GWAS, and general risk tolerance is moderately to strongly genetically correlated ([Formula] to 0.50) with a range of risky behaviors. Bioinformatics analyses imply that genes near general-risk-tolerance-associated SNPs are highly expressed in brain tissues and point to a role for glutamatergic and GABAergic neurotransmission. We find no evidence of enrichment for genes previously hypothesized to relate to risk tolerance.

genetics↗

bioRxiv2018

Identification of candidate genes for gelatinization temperature, gel consistency and pericarp color by GWAS in rice based on SLAF-sequencing

Rice is an important cereal in the world, uncovering the genetic basis of agronomic traits in rice landraces genes associated with agronomically important traits is indispensable for both understanding the genetic basis of phenotypic variation and efficient crop improvement. Gelatinization temperature, gel consistency and pericarp color are important indices of rice cooking and eating quality evaluation and potential nutritional importance, which attract wide attentions in the application of genetic and breeding. To dissect the genetic basis of gelatinization temperature (GT), gel consistency (GC) and pericarp color (PC), a total of 419 rice landraces core germplasm collections consisting of 330 indica lines, 78 japonica lines and 11 uncertain varieties were grown, collected, then GT, GC, PC were measured for two years, and sequenced using Specific Locus Amplified Fragment Sequencing (SLAF) technology. In this study, 261,385,070 clean reads and 56,768 polymorphic SLAF tags were obtained, which a total of 211,818 single nucleotide polymorphisms (SNPs) were discovered. With 208,993 SNPs meeting the criterion of minor allele frequency (MAF) > 0.05 and integrity> 0.5, the phylogenetic tree and population structure analysis were performed for all 419 rice landraces, and the whole panel mainly separated into six subpopulations based on population structure analysis. Genome-wide association study (GWAS) was carried out for the whole panel, indica subpanel and japonica subpanel with subset SNPs respectively. One quantitative trait locus (QTL) on chromosome 6 for GT was detected in the whole panel and indica subpanel, and one QTL associated with GC was located on chromosome 6 in the whole panel and indica subpanel. For the PC trait, 8 QTLs were detected in the whole panel on chromosome 1, 3, 4, 7, 8, 10 and 11, and 7 QTLs in the indica subpanel on chromosome 3, 4, 7, 8, 10 and 11. The loci on chromosome 3, 8, 10 and 11 have not been identified previously, and they may be the candidate genes of pericarp color. For the three traits, no QTL was detected in japonica subpanel probably because of the polymorphism repartition between the subpanel, or small population size of japonica subpanel. This paper provides new gene resources and insights into the molecular mechanisms of important agricultural trait of rice phenotypic variation and genetic improvement of rice quality variety breeding.

genomics↗

bioRxiv2017

Improved Aedes aegypti mosquito reference genome assembly enables biological discovery and vector control

Female Aedes aegypti mosquitoes infect hundreds of millions of people each year with dangerous viral pathogens including dengue, yellow fever, Zika, and chikungunya. Progress in understanding the biology of this insect, and developing tools to fight it, has been slowed by the lack of a high-quality genome assembly. Here we combine diverse genome technologies to produce AaegL5, a dramatically improved and annotated assembly, and demonstrate how it accelerates mosquito science and control. We anchored the physical and cytogenetic maps, resolved the size and composition of the elusive sex-determining \"M locus\", significantly increased the known members of the glutathione-S-transferase genes important for insecticide resistance, and doubled the number of chemosensory ionotropic receptors that guide mosquitoes to human hosts and egg-laying sites. Using high-resolution QTL and population genomic analyses, we mapped new candidates for dengue vector competence and insecticide resistance. We predict that AaegL5 will catalyse new biological insights and intervention strategies to fight this deadly arboviral vector.

genomics↗

bioRxiv2017

Clustering enzymes using E.coli inner cell membrane as scaffold in metabolic pathway

Clustering enzymes in the same metabolism pathway is a natural strategy to enhance the productivity. Several systems have been designed to artificially cluster desired enzymes in the cell, such as synthetic protein scaffold and nucleic acid scaffold. However, these scaffolds require complicated construction process and have limited slots for target enzymes. Following this direction, we designed a scaffold system based on natural cell membrane. Target enzymes (FabZ, FabG, FabI and TesA in fatty acid synthesis II pathway) are anchored on the E.coli inner membrane, showing the enhanced metabolism flux without the requirement of the further artificial interactions to force the clustering. Furthermore, anchoring the enzymes on the membrane enhances the products exportation, which further increases the productivity. Together, the proposed system has potential applications in producing valuable biomaterials.

synthetic biology↗

bioRxiv2017

Phytophthora methylomes modulated by expanded 6mA methyltransferases are associated with adaptive genome regions

Filamentous plant pathogen genomes often display a bipartite architecture with gene sparse, repeat-rich compartments serving as a cradle for adaptive evolution. However, the extent to which this \"two-speed\" genome architecture is associated with genome-wide epigenetic modifications is unknown. Here, we show that the oomycete plant pathogens Phytophthora infestans and Phytophthora sojae possess functional adenine N6- methylation (6mA) methyltransferases that modulate patterns of 6mA marks across the genome. In contrast, 5-methylcytosine (5mC) could not be detected in the two Phytophthora species. Methylated DNA IP Sequencing (MeDIP-seq) of each species revealed that 6mA is depleted around the transcriptional starting sites (TSS) and is associated with low expressed genes, particularly transposable elements. Remarkably, genes occupying the gene-sparse regions have higher levels of 6mA compared to the remainder of both genomes, possibly implicating the methylome in adaptive evolution of Phytophthora. Among three putative adenine methyltransferases, DAMT1 and DAMT3 displayed robust enzymatic activities. Surprisingly, single knockouts of each of the 6mA methyltransferases in P. sojae significantly reduced in vivo 6mA levels, indicating that the three enzymes are not fully redundant. MeDIP-seq of the damt3 mutant revealed uneven patterns of 6mA methylation across genes, suggesting that PsDAMT3 may have a preference for gene body methylation after the TSS. Our findings provide evidence that 6mA modification is an epigenetic mark of Phytophthora genomes and that complex patterns of 6mA methylation by the expanded 6mA methyltransferases may be associated with adaptive evolution in these important plant pathogens.

molecular biology↗

bioRxiv2017

A fine-tuned vector-parasite dialogue in tsetse’s cardia determines peritrophic matrix integrity and trypanosome transmission success

Arthropod vectors have multiple physical and immunological barriers that impede the development and transmission of parasites to new vertebrate hosts. These barriers include the peritrophic matrix (PM), a chitinous barrier that separates the blood bolus from the midgut epithelia and inturn, modulates vector-microbiota interactions. In tsetse flies, a sleeve-like PM is continuously produced by the cardia organ located at the fore- and midgut junction. African trypanosomes, Trypanosoma brucei, must bypass the PM twice; first to colonize the midgut and secondly to reach the salivary glands (SG), to complete their transmission cycle in tsetse. However, not all flies with midgut infections develop mammalian transmissible SG infections - the reasons for which are unclear. Here, we used transcriptomics, microscopy and functional genomics analyses to understand the factors that regulate parasite migration from midgut to SG. In flies with midgut infections only, parasites fail to cross the PM as they are eliminated from the cardia by reactive oxygen intermediates (ROIs) - albeit at the expense of collateral cytotoxic damage to the cardia. In flies with midgut and SG infections, expression of genes encoding components of the PM is reduced in the cardia, and structural integrity of the PM barrier is compromised. Under these circumstances trypanosomes traverse through the newly secreted and compromised PM. The process of PM attrition that enables the parasites to re-enter into the midgut lumen is apparently mediated by components of the parasites residing in the cardia. Thus, a fine-tuned dialogue between tsetse and trypanosomes at the cardia determines the outcome of PM integrity and trypanosome transmission success.\n\nAuthor summaryInsects are responsible for transmission of parasites that cause deadly diseases in humans and animals. Understanding the key factors that enhance or interfere with parasite transmission processes can result in new control strategies. Here, we report that a proportion of tsetse flies with African trypanosome infections in their midgut can prevent parasites from migrating to the salivary glands, albeit at the expense of collateral damage. In a subset of flies with gut infections, the parasites manipulate the integrity of a midgut barrier, called the peritrophic matrix, and reach the salivary glands for transmission to the next mammal. Either targeting parasite manipulative processes or enhancing peritrophic matrix integrity could reduce parasite transmission.

microbiology↗

bioRxiv2017

Assessment of the need for separate test set and number of medical images necessary for deep learning: a sub-sampling study

Deep learning algorithms have tremendous potential utility in the classification of biomedical images. For example, images acquired with retinal optical coherence tomography (OCT) can be used to accurately classify patients with adult macular degeneration (AMD), and distinguish them from healthy control patients. However, previous research has suggested that large amounts of data are required in order to train deep learning algorithms, because of the large number of parameters that need to be fit. Here, we show that a moderate amount of data (data from approximately 1,800 patients) may be enough to reach close-to-maximal performance in the classification of AMD patients from OCT images. These results suggest that deep learning algorithms can be trained on moderate amounts of data, provided that images are relatively homogenous, and the effective number of parameters is sufficiently small. Furthermore, we demonstrate that in this application, cross-validation with a separate test set that is not used in any part of the training does not differ substantially from cross-validation with a validation data-set used to determine the optimal stopping point for training.

neuroscience↗

bioRxiv2017

PrimerServer: a high-throughput primer design and specificity-checking platform

SummaryDesigning specific primers for multiple sites across the whole genome is still challenging, especially in species with complex genomes. Here we present PrimerServer, a high-throughput primer design and specificity-checking platform with both web and command-line interfaces. This platform efficiently integrates site selection, primer design, specificity checking and data presentation. In our case study, PrimerServer achieved high accuracy and a fast running speed for a large number of sites, suggesting its potential for molecular biology applications such as molecular breeding or medical testing.\n\nAvailability and ImplementationSource code for PrimerServer is available at https://github.com/billzt/PrimerServer. A demo server is freely accessible at https://primerserver.org, with all major browsers supported.\n\nContactzhangrui@caas.cn or guosandui@caas.cn

bioinformatics↗