Search bioRxiv⌕ Search

Biology subjects

Zhu, W.

Publications and source records attributed to Zhu, W..

At least 181 records · Page 10Linked to original sources

Trehalose Stabilizing Protein in a Water Replacement Scenario: Insights from Molecular Dynamics Simulation

How trehalose has exceptional property in helping biomolecules preserve their native structures remains a subject of active research. Running molecular dynamics simulations on a model protein in low-concentrated trehalose solution and pure water, respectively, the present study verifies the ability of trehalose in stabilizing protein native structure and provides a comprehensive atomic-level picture of the molecular interactions among protein, trehalose, and water in their mixed solution. Trehalose directly interacts to and meanwhile affects the interactions between the other species via hydrogen bonding: 1) trehalose molecules are clustered through inter-molecular hydrogen bonding interaction; 2) trehalose forms hydrogen bond with water which influences the strength of water-water hydrogen bonding network but does not impair protein-water hydrogen bonding; 3) trehalose is accessible to form hydrogen bonds towards protein and simultaneously replace water molecules around protein which reduces the hydrogen bonding possibility from water to protein, in accordance with "water replacement" scenario.

biophysics↗

Local nucleosome dynamics and eviction following a double-strand break are reversible by NHEJ-mediated repair in the absence of DNA replication

Although the molecular events required for the repair of double-strand breaks (DSB) have been well characterized, the role of epigenetic processes in the recognition and repair of DSBs has only been investigated at low resolution. We rapidly and synchronously induced a site-specific DSB in Saccharomyces cerevisiae upstream of the PHO5 locus, which has well-positioned nucleosomes. Utilizing MNase-seq epigenome mapping we interrogated the order of chromatin changes that occur immediately following a DSB by generating a base-pair resolution map of the chromatin landscape. In wild type cells, the first nucleosome left of the break was rapidly evicted. The eviction of this flanking nucleosome was dynamic and proceeded through an early intermediate chromatin structure where the nucleosome was repositioned in the adjacent linker DNA. Other nucleosomes bordering both sides of the break were also shifted away from the break; however, their loss was more gradual. These local changes preceded a broader loss of chromatin organization and nucleosome eviction that was marked by increased MNase sensitivity in the regions [~]8 kb on each side of the break. While the broad loss of chromatin organization was dependent on the end-processing complex, Mre11-Rad50-Xrs2 (MRX), the early remodeling and repositioning of the nucleosome adjacent to the break was independent of the MRX and YKU70/80 complexes. We also examined the temporal dynamics of NHEJ-mediated repair in a G1-arrested population, where 5 to 3 end-resection of DSB ends is blocked. Concomitant with DSB repair, we observed the re-deposition and precise re-positioning of nucleosomes at the originally-occupied positions. This re-establishment of the pre-lesion chromatin landscape suggests that a DNA replication-independent mechanism exists in G1 cells to preserve epigenome organization following DSB repair.

genomics↗

Phoenix Enhancer: an online service/tool for proteomics data mining using clustered spectra

MotivationSpectrum clustering has been used to enhance proteomics data analysis: some originally unidentified spectra can potentially be identified and individual peptides can be evaluated to find potential mis-identifications by using clusters of identified spectra. The Phoenix Enhancer provides an infrastructure to analyze tandem mass spectra and the corresponding peptides in the context of previously identified public data. Based on PRIDE Cluster data and a newly developed pipeline, four functionalities are provided: i) evaluate the original peptide identifications in an individual dataset, to find low confidence peptide spectrum matches (PSMs) which could correspond to mis-identifications; ii) provide confidence scores for all originally identified PSMs, to help users evaluate their quality (complementary to getting a global false discovery rate); iii) identify potential new PSMs for originally unidentified spectra; and iv) provide a collection of browsing and visualization tools to analyze and export the results. In addition to the web based service, the code is open-source and easy to re-deploy on local computers using Docker containers. AvailabilityThe service of Phoenix Enhancer is available at http://enhancer.ncpsb.org. All source code is freely available in GitHub (https://github.com/phoenix-cluster/) and can be deployed in the Cloud and HPC architectures. Contactbaimz@cqupt.edu.cn Supplementary informationSupplementary data are available online.

bioinformatics↗

Structure and mechanism of a Type III CRISPR defence DNA nuclease activated by cyclic oligoadenylate

The CRISPR system provides adaptive immunity against mobile genetic elements in prokaryotes. On binding invading RNA species, Type III CRISPR systems generate cyclic oligoadenylate (cOA) molecules which act as a second messenger, signalling infection and potentiating a powerful immune response by activating a range of downstream effector proteins that can lead to viral clearance, cell dormancy or death. Only one type of effector enzyme has been studied - the Csm6/Csx1 ribonuclease domain, and the mechanism of cOA activation is not understood at a molecular level. Here we describe the structure and mechanism of a novel cOA-activated CRISPR defence DNA endonuclease, Can1 (\"CRISPR ancillary nuclease 1\"). Can1 has a unique monomeric structure with two CRISPR associated Rossman fold (CARF) CARF domains and two DNA nuclease-like domains. The crystal structure of the enzyme has been captured in the activated state, with a cyclic tetra-adenylate (cA4) molecule bound at the core of the protein. cA4 binding reorganises the structure to license a metal-dependent DNA nuclease activity specific for nicking of supercoiled DNA. DNA nicking by Can1 is predicted to slow down viral replication kinetics by leading to the collapse of DNA replication forks.

biochemistry↗

Satellite Imaging of Global Urbanicity relate to Adolescent Brain Development and Behavior

Urbanicity, the impact of living in urban areas, is among the greatest environmental challenges for mental health. While urbanicity might be distinct in different sociocultural conditions and geographic locations, there are likely to exist common features shared in different areas of the globe. Understanding these common and specific relations of urbanicity with human brain and behavior will enable to assess the impact of urbanicity on mental disorders, especially in childhood and adolescence, where prevention and early interventions are likely to be most effective.\n\nWe constructed from satellite-based remote sensing data a factor for urbanicity that was highly correlated with population density ground data. This factor, UrbanSat was utilized in the Chinese CHIMGEN sample (N=831) and the longitudinal European IMAGEN cohort (N=810) to investigate if exposure to urbanicity during childhood and adolescence is associated with differences in brain structure and function in young adults, and if these changes are linked to behavior.\n\nUrbanicity was found negatively correlated with medial prefrontal cortex volume and positively correlated with cerebellar vermis volume in young adults from both China and Europe. We found an increased correlation of urbanicity with functional network connectivity within- and between- brain networks in Chinese compared to European participants. Urbanicity was highly correlated with a measure of perceiving a situation from the perspective of others, as well as symptoms of depression in both datasets. These correlations were mediated by the structural and functional brain changes observed. Susceptibility to urbanicity was greatest in two developmental windows during mid-childhood and adolescence.\n\nUsing innovative technology, we were able to probe the relationship between urban upbringing with brain change and behavior in different sociocultural conditions and geographic locations. Our findings help to identify shared and distinct determinants of adolescent brain development and mental health in different regions of the world, thus contributing to targeted prevention and early-intervention programs for young people in their unique environment. Our approach may be relevant for public health, policy and urban planning globally.

neuroscience↗

Substrate Transport is Mediated not only by P-glycoprotein but also by Lipid Penetration

In association with large-scale conformational changes, the members of the ATP-binding cassette (ABC) transporter superfamily such as P-glycoprotein (P-gp) pump endogenous cytotoxic substances and exogenous drugs out of cells. Here, a series of nonequilibrium-driven molecular dynamics (MD) simulations are sophisticatedly combined to provide a generally effective access to quantitatively investigate such a complex biological process that has been posing a great challenge for experiments and computational simulations. Both common features and unique characteristics of multiple ligands (substrates or inhibitors) that are recognized by P-gps from mouse and human species are quantitatively explored, providing additional insights into experimentally suggested ligand transport pathways and summarizing the important roles of not only different P-gps but also lipids in regulating ligand transport. These findings reveal the molecular mechanism underlying the transport of ligands by P-gps from different species and emphasize the consideration of lipid effects on the future design of effective P-gp inhibitors.

biophysics↗

Neural Multi-Scale Self-Supervised Registration for Echocardiogram Dense Tracking

Echocardiography has become routinely used in the diagnosis of cardiomyopathy and abnormal cardiac blood flow. However, manually measuring myocardial motion and cardiac blood flow from echocar-diogram is time-consuming and error-prone. Computer algorithms that can automatically track and quantify myocardial motion and cardiac blood flow are highly sought after, but have not been very successful due to noise and high variability of echocardiography. In this work, we propose a neural multi-scale self-supervised registration (NMSR) method for automated myocardial and cardiac blood flow dense tracking. NMSR incorporates two novel components: 1) utilizing a deep neural net to parameterize the velocity field between two image frames, and 2) optimizing the parameters of the neural net in a sequential multi-scale fashion to account for large variations within the velocity field. Experiments demonstrate that NMSR yields significantly better registration accuracy than the state-of-the-art methods, such as advanced normalization tools (ANTs) and Voxel Morph, for both myocardial and cardiac blood flow dense tracking. Our approach promises to provide a fully automated method for fast and accurate analyses of echocardiograms.

bioengineering↗

Precision Genome Editing Using Bacteriophage-derived Peptides

BackgroundCRISPR-Cas9 has been developed as a therapeutic agent for various infectious and genetic diseases. In many clinically relevant applications, constitutively active CRISPR-Cas9 is delivered into human cells without a temporal control system. Excessive and prolonged expression CRISPR-Cas9 can lead to elevated off-target cleavage. The need for modulating CRISPR-Cas9 activity over the dimensions of time and dose has created the demand of developing CRISPR-Cas off-switches. Protein and small molecule-based CRISPR-Cas inhibitors have been reported in previous studies. ResultsWe report the discovery of Cas9-inhibiting peptides from inoviridae bacteriophages. These peptides, derived from the periplasmic domain of phage major coat protein G8P (G8PPD), can inhibit the in vitro activity of Streptococcus pyogenes Cas9 (SpCas9) proteins in an allosteric manner. Importantly, the inhibitory activity of G8PPD on SpCas9 is dependent on the order of guide RNA addition. Ectopic expression of full-length G8P (G8PFL) or G8PPD in human cells can inactivate the genome-editing activity of SpCas9 with minimum alterations of the mutation patterns. Furthermore, unlike the anti-CRISPR protein AcrII4A that completely abolishes the cellular activity of CRISPR-Cas9, G8P co-transfection can reduce the off-target activity of co-transfected SpCas9 while retaining its on-target activity. ConclusionG8Ps discovered in the current study represent the first anti-CRISPR peptides that can allosterically inactivate CRISPR-Cas9. This finding may provide insights into developing next-generation CRISPR-Cas inhibitors for precision genome engineering.

genomics↗

Molecular basis of human asparagine synthetase inhibitor specificity

Expression of the enzyme human asparagine synthetase (ASNS) promotes metastatic progression in breast cancer, which affects L-asparagine levels and tumor cell invasiveness. Human ASNS has therefore emerged as a bona fide drug target for cancer therapy. We have reported a slow-onset, tight binding ASNS inhibitor with nanomolar affinity, but our compound exhibits poor cell permeability. On the other hand, we show here that this inhibitor exhibits remarkable selectivity for the human ASNS in HCT-116 cell lysates. By determining the first high-resolution (1.85 [A]) X-ray crystal structure for human ASNS, we have built a computational model of the enzyme complexed to our inhibitor, which provides the first insights into the intermolecular interactions mediating specificity. These findings should facilitate the development of a second generation of ASNS inhibitors, leading to the discovery of drugs to prevent metastasis.

biochemistry↗

Identification of genome-wide nucleotide sites associated with mammalian virulence in influenza A viruses

MotivationThe virulence of influenza viruses is a complex multigenic trait. Previous studies about the virulence determinants of influenza viruses mainly focused on amino acid sites, ignoring the influence of nucleotide mutations.\n\nResultsWe collected more than 200 viral strains from 21 subtypes of influenza A viruses with virulence in mammals and obtained over 100 mammalian virulence-related nucleotide sites across the genome by computational analysis. Interestingly, 50 of these nucleotide sites only experienced synonymous mutations. Further experiments showed that synonymous mutations in the top two of these nucleotide sites, i.e., PB1-2031 and PB1-633, enhanced the pathogenicity of the viruses in mice. Finally, machine-learning models with accepted accuracy for predicting mammalian virulence of influenza A viruses were built. Overall, this study highlighted the importance of nucleotide mutations, especially synonymous mutations in viral virulence, and provided rapid methods for evaluating the virulence of influenza A viruses. It could be helpful for early warning of newly emerging influenza A viruses.

microbiology↗

Human antibody cocktail deploys multiple functions to confer pan-ebolavirus protection

During the unprecedented 2013-2016 Ebola virus disease (EVD) epidemic in Western Africa and in its aftermath, the passive administration of monoclonal antibodies (mAbs) emerged as a promising treatment approach1-7. However, all antibody-based therapeutics currently in advanced development are specific for a single member of the Ebolavirus genus, Ebola virus (EBOV), and ineffective against divergent outbreak-causing ebolaviruses, including Bundibugyo virus (BDBV) and Sudan virus (SUDV)2,3,5,7. Here we advance MBP134, a cocktail of two broadly neutralizing human mAbs targeting the filovirus surface glycoprotein, GP, as a candidate pan-ebolavirus therapeutic. One component of this cocktail is a pan-ebolavirus neutralizing mAb, ADI-15878, isolated from a human EVD survivor8,9. The second, ADI-23774, was derived by affinity maturation of a human mAb8,9 via yeast display to enhance its potency against SUDV. MBP134 afforded exceptionally potent pan-ebolavirus neutralization in vitro and demonstrated greater protective efficacy than ADI-15878 alone in the guinea pig model of lethal EBOV challenge. A second-generation cocktail, MBP134AF, engineered to effectively harness natural killer (NK) cells afforded additional, unprecedented improvements in protective efficacy against EBOV and SUDV in guinea pigs relative to both its precursor and to any mAbs or mAb cocktails tested previously. MBP134AF is a best-in-class mAb cocktail suitable for evaluation as a pan-ebolavirus therapeutic in nonhuman primates.

microbiology↗

Two-antibody pan-ebolavirus cocktail confers broad therapeutic protection in ferrets and nonhuman primates

All available experimental vaccines and immunotherapeutics1,2 against Ebola virus (EBOV), including rVSV-ZEBOV3 and ZMappTM4, lack activity against other ebolaviruses associated with human disease outbreaks. This year, two separate outbreaks of EBOV in the Democratic Republic of Congo underscored the unpredictable nature of ebolavirus reemergence in a region that has historically experienced outbreaks of the divergent ebolaviruses Sudan virus (SUDV) and Bundibugyo virus (BDBV)5. Here we show that MBP134AF, a pan-ebolavirus therapeutic comprising two broadly neutralizing human antibodies (bNAbs)6,7(see companion manuscript, Wec et al.) could protect against lethal EBOV, SUDV, and BDBV infection in ferrets and nonhuman primates (NHPs). MBP134AF not only not only establishes a viable therapeutic countermeasure to outbreaks caused by antigenically diverse ebolaviruses but also affords unprecedented effectiveness and potency--a single 25-mg/kg dose was fully protective in NHPs. This best-in-class antibody cocktail is the culmination of an intensive collaboration spanning academia, industry and government in response to the 2013-2016 EBOV epidemic6,7 and provides a translational research model for the rapid development of immunotherapeutics targeting emerging infectious diseases.

microbiology↗

AnatomyNet: Deep 3D Squeeze-and-excitation U-Nets for fast and fully automated whole-volume anatomical segmentation

PurposeRadiation therapy (RT) is a common treatment for head and neck (HaN) cancer where therapists are often required to manually delineate boundaries of the organs-at-risks (OARs). The radiation therapy planning is time-consuming as each computed tomography (CT) volumetric data set typically consists of hundreds to thousands of slices and needs to be individually inspected. Automated head and neck anatomical segmentation provides a way to speed up and improve the reproducibility of radiation therapy planning. Previous work on anatomical segmentation is primarily based on atlas registrations, which takes up to hours for one patient and requires sophisticated atlas creation. In this work, we propose the AnatomyNet, an end-to-end and atlas-free three dimensional squeeze-and-excitation U-Net (3D SE U-Net), for fast and fully automated whole-volume HaN anatomical segmentation.\n\nMethodsThere are two main challenges for fully automated HaN OARs segmentation: 1) challenge in segmenting small anatomies (i.e., optic chiasm and optic nerves) occupying only a few slices, and 2) training model with inconsistent data annotations with missing ground truth for some anatomical structures because of different RT planning. We propose the AnatomyNet that has one down-sampling layer with the trade-off between GPU memory and feature representation capacity, and 3D SE residual blocks for effective feature learning to alleviate these challenges. Moreover, we design a hybrid loss function with the Dice loss and the focal loss. The Dice loss is a class level distribution loss that depends less on the number of voxels in the anatomy, and the focal loss is designed to deal with highly unbalanced segmentation. For missing annotations, we propose masked loss and weighted loss for accurate and balanced weights updating in the learning of the AnatomyNet.\n\nResultsWe collect 261 HaN CT images to train the AnatomyNet, and use MICCAI Head and Neck Auto Segmentation Challenge 2015 as the benchmark dataset to evaluate the performance of the AnatomyNet. The objective is to segment nine anatomies: brain stem, chiasm, mandible, optic nerve left, optic nerve right, parotid gland left, parotid gland right, submandibular gland left, and submandibular gland right. Compared to previous state-of-the-art methods for each anatomy from the MICCAI 2015 competition, the AnatomyNet increases Dice similarity coefficient (DSC) by 3.3% on average. The proposed AnatomyNet takes only 0.12 seconds on average to segment a whole-volume HaN CT image of an average dimension of 178 x 302 x 225. All the data and code will be availablea.\n\nConclusion1We propose an end-to-end, fast and fully automated deep convolutional network, AnatomyNet, for accurate and whole-volume HaN anatomical segmentation. The proposed Anato-myNet outperforms previous state-of-the-art methods on the benchmark dataset. Extensive experiments demonstrate the effectiveness and good generalization ability of the components in the AnatomyNet.

bioengineering↗

DeepEM: Deep 3D ConvNets With EM For Weakly Supervised Pulmonary Nodule Detection

Recently deep learning has been witnessing widespread adoption in various medical image applications. However, training complex deep neural nets requires large-scale datasets labeled with ground truth, which are often unavailable in many medical image domains. For instance, to train a deep neural net to detect pulmonary nodules in lung computed tomography (CT) images, current practice is to manually label nodule locations and sizes in many CT images to construct a sufficiently large training dataset, which is costly and difficult to scale. On the other hand, electronic medical records (EMR) contain plenty of partial information on the content of each medical image. In this work, we explore how to tap this vast, but currently unexplored data source to improve pulmonary nodule detection. We propose DeepEM, a novel deep 3D ConvNet framework augmented with expectation-maximization (EM), to mine weakly supervised labels in EMRs for pulmonary nodule detection. Experimental results show that DeepEM can lead to 1.5% and 3.9% average improvement in free-response receiver operating characteristic (FROC) scores on LUNA16 and Tianchi datasets, respectively, demonstrating the utility of incomplete information in EMRs for improving deep learning algorithms.1

bioinformatics↗

Modulation of ACD6 dependent hyperimmunity by natural alleles of an Arabidopsis thaliana NLR resistance gene

Plants defend themselves against pathogens by activating an array of immune responses. Unfortunately, immunity programs may also cause unintended collateral damage to the plant itself. The quantitative disease resistance gene ACCELERATED CELL DEATH 6 (ACD6) serves as a nexus for the trade-off between growth and pathogen resistance in wild populations of Arabidopsis thaliana. An autoimmune allele, ACD6-Est, first identified in the natural accession Est-1, is found in over 10% of wild strains, even though it causes a clear fitness penalty under optimal growth conditions. There is, however, extensive variation in the strength of the autoimmune phenotype expressed by strains with an ACD6-Est allele, indicative of genetic modifiers. Quantitative genetic analysis suggests that the population genetic basis of ACD6 modulation is complex, with different strains often carrying different large-effect modifiers. One modifier is SUPPRESSOR OF NPR1-1, CONSTITUTIVE 1 (SNC1), located in a highly polymorphic cluster of nucleotide-binding domain and leucine-rich repeat (NLR) immune receptor genes, which are prototypes for qualitative disease resistance genes. Allelic variation at SNC1 correlates with ACD6-Est activity in multiple accessions, and a common structural variant affecting the NL linker sequence can explain differences in SNC1 activity. Taken together, we find that an NLR gene can mask the activity of an ACD6 autoimmune allele in natural A. thaliana populations, thereby linking different arms of the plant immune system.\n\nAuthor summaryPlants defend themselves against pathogens by activating immune responses. Unfortunately, these can cause unintended collateral damage to the plant itself. Nevertheless, some wild plants have genetic variants that confer a low threshold for the activation of immunity. While these enable a plant to respond particularly quickly to pathogen attack, such variants might be potentially dangerous. We are investigating one such variant of the immune gene ACCELERATED CELL DEATH 6 (ACD6) in the plant Arabidopsis thaliana. We discovered that there are variants at other genetic loci that can mask the effects of an overly active ACD6 gene. One of these genes, SUPPRESSOR OF NPR1-1, CONSTITUTIVE 1 (SNC1), codes for a known immune receptor. The SNC1 variant that attenuates ACD6 activity is rather common in A. thaliana populations, suggesting that new combinations of the hyperactive ACD6 variant and this antagonistic SNC1 variant will often arise by natural crosses. Similarly, because the two genes are unlinked, outcrossing will often lead to the hyperactive ACD6 variants being unmasked again. We propose that allelic diversity at SNC1 contributes to the maintenance of the hyperactive ACD6 variant in natural A. thaliana populations.

plant biology↗

The plasma miR-122 basal levels respond to circulating catecholamine in rats

miR-122 in circulation is a promising non-invasive biomarker as a replacement or supplement of current serum biomarkers for liver injuries. But the concept was questioned by recent studies, mainly due to its release from hepatocytes in absence of overt cellular injuries. In this study, we reported that the hepatic metabolism of circulating catecholamines resulted in the release of hepatocyte-specific miR-122. Acute stress-induced hepatocellular deformation was histopathologically different from drug-induced liver injury with significant increases of plasma miR-122 levels. The basal levels of human plasma miR-122 could be significantly altered by emotional responses. Interday variances of plasma miR-122 measurements were reduced effectively by stress-relief measures. The metabolism of basal circulating norepinephrine and epinephrine in liver might contribute to the basal levels of plasma miRNAs expressed in hepatocytes.

molecular biology↗

The brown alga Saccharina japonica features distinct vanadium-dependent bromoperoxidases and iodoperoxidases

Marine algae have an extraordinary ability to absorb halogens which provide algae with an inorganic antioxidant impacting atmospheric chemistry. Although brown algal Laminariales species are the most efficient iodine accumulators among all living systems, and Saccharina japonica is the primary material used for iodine extraction, the functions and regulatory mechanisms of these species have not been fully documented. In this study, a functional genomics analysis of the algal vanadium-dependent haloperoxidase (vHPO) gene family was conducted; there genes can introduce halogen atoms into organic compounds. The comprehensive analyses regarding the bioinformatics and phylogenetics of novel genomic and transcriptomic sequencing data of 21 Rhodophyta and 19 Ochrophyta marine algal species revealed that brown algal vHPOs have two gene types, vanadium-dependent bromoperoxidase (vBPO) and vanadium-dependent iodoperoxidase (vIPO), with secondary endosymbiotic host origin. The enzyme activity of S. japonica vBPO and vIPO were verified for the first time and were quite stable in a wide range of temperature and pH values. However, the specific activity and optimal conditions were considerably different between vBPO and vIPO. The transcript expression analysis in different S. japonica tissues (including rhizoids), generations (sporophytes and gametophytes), sexes (male and female), and stress conditions (hyposaline and hyperthermia) also showed great differences between vBPOs and vIPOs. Most of the vBPOs were constitutively expressed with higher expression dose, which may be responsible for basal halogen metabolism. On the contrary, vIPOs mainly showed specific expression, which may be involved in tissue differentiation, generation differentiation, sex differentiation, and stress regulation. Comprehensive analysis of gene family evolution, enzyme biochemical characteristics, and complex transcriptional mechanisms were conducive to the environmental adaptation and sophisticated system evolution of Laminariales. The successful bromination of small-molecule compound substrate by SjavBPO provided high activity and efficient enzymatic tools for artificial synthesis of halogenated compounds.

molecular biology↗

Improving the calling of non-invasive prenatal testing on 13-/18-/21-trisomy by support vector machine discrimination

With the advance of next-generation sequencing technologies, non-invasive prenatal testing (NIPT) has been developed and employed in fetal aneuploidy screening on 13-/18-/21-trisomies through detecting cell-free fetal DNA (cffDNA) in maternal blood. Although Z test is widely used in NIPT nowadays, there is still necessity to improve its accuracy for removing a) false negatives and false positives, and b) the ratio of unclassified data, so as to reduce the potential harm to patients caused by these inaccuracies as well as the induced cost of retests.\n\nEmploying multiple Z tests with machine-learning algorithm could provide a better prediction on NIPT data. Combining the multiple Z values with indexes of clinical signs and quality control, features were collected from the known samples and scaled for model training in support vector machine (SVM) discrimination. The trained model was applied to predict the unknown samples, which showed significant improvement. In 4752 qualified NIPT data, our method reached 100% accuracies on all three chromosomes, including 151 data that were grouped as unclassified by one-Z-value based method. Moreover, four false positives and four false negatives were corrected by using this machine-learning model.\n\nTo our knowledge, this is the first study to employ support vector machine in NIPT data analysis. It is expected to replace the current one-Z-value based NIPT analysis in clinical use.

bioinformatics↗