Biology subjects

Dai, Y.

Publications and source records attributed to Dai, Y..

At least 145 records · Page 8Linked to original sources

bioRxiv2021

Drug-driven reclassification of multiple tumour subtypes reveals intrinsic molecular concordance of therapy across histologically disparate cancers

Cancers that are histologically defined as the same type of cancer often need a distinct therapy based on underlying heterogeneity; likewise, histologically disparate cancers can require similar treatment approaches due to intrinsic similarities. A comprehensive analysis integrated with drug response data and genomic alterations, particularly to reveal therapeutic concordance mechanisms across histologically disparate tumour subtypes, has not yet been fully exploited. In this study, we used pharmacogenomic profiling data provided from the Cancer Genome Project (CGP) in a systematic in silico investigation of the pharmacological subtypes of cancers and the intrinsic concordance of molecular mechanisms leading to similar therapeutic responses across histologically disparate tumour subtypes. We further developed a novel approach to redefine cell-to-cell similarity and drug-to-drug similarity from the therapeutic concordance, providing a new point of view to study cancer heterogeneity. Our study identified that histologically different tumours, such as malignant melanoma and colorectal adenocarcinoma, could belong to the same pharmacological subtype regarding drug sensitivity to MEK inhibitors, which was determined by their genomic alterations, high occurrence of BRAF or KRAS mutations. Therapeutic concordance for chemotherapy drugs was identified across histologically disparate hematological tumors mainly due to the extraordinary activation of the cell cycle in blood cancers. A subcluster of SCLC had a more similar profile with hematological tumors, and was associated with the malignant phenotype, with a higher level of MYC expression. We developed a website to store and visualize the pharmacological subtypes of drugs, as well as their connected genomic and expression alterations.

bioinformatics↗

bioRxiv2021

Association of CXCR6 with COVID-19 severity: Delineating the host genetic factors in transcriptomic regulation

BackgroundThe coronavirus disease 2019 (COVID-19) is an infectious disease that mainly affects the host respiratory system with [~]80% asymptomatic or mild cases and [~]5% severe cases. Recent genome-wide association studies (GWAS) have identified several genetic loci associated with the severe COVID-19 symptoms. Delineating the genetic variants and genes is important for better understanding its biological mechanisms. MethodsWe implemented integrative approaches, including transcriptome-wide association studies (TWAS), colocalization analysis and functional element prediction analysis, to interpret the genetic risks using two independent GWAS datasets in lung and immune cells. To understand the context-specific molecular alteration, we further performed deep learning-based single cell transcriptomic analyses on a bronchoalveolar lavage fluid (BALF) dataset from moderate and severe COVID-19 patients. ResultsWe discovered and replicated the genetically regulated expression of CXCR6 and CCR9 genes. These two genes have a protective effect on the lung and a risk effect on whole blood, respectively. The colocalization analysis of GWAS and cis-expression quantitative trait loci highlighted the regulatory effect on CXCR6 expression in lung and immune cells. In the lung resident memory CD8+ T (TRM) cells, we found a 3.32-fold decrease of cell proportion and lower expression of CXCR6 in the severe than moderate patients using the BALF transcriptomic dataset. Pro-inflammatory transcriptional programs were highlighted in TRM cells trajectory from moderate to severe patients. ConclusionsCXCR6 from the 3p21.31 locus is associated with severe COVID-19. CXCR6 tends to have a lower expression in lung TRM cells of severe patients, which aligns with the protective effect of CXCR6 from TWAS analysis. We illustrate one potential mechanism of host genetic factor impacting the severity of COVID-19 through regulating the expression of CXCR6 and TRM cell proportion and stability. Our results shed light on potential therapeutic targets for severe COVID-19.

genetics↗

bioRxiv2021

REAP: A platform to identify autoantibodies that target the human exoproteome

Autoantibodies that recognize extracellular proteins (the "exoproteome") exert potent biological effects but have proven challenging to detect with existing screening technologies. Here, we developed Rapid Extracellular Antigen Profiling (REAP) as a technique for comprehensive, high-throughput discovery of exoproteome-targeting autoantibodies. With REAP, patient samples are applied to a genetically-barcoded library containing 2,688 human extracellular proteins displayed on the surface of yeast. Antibody-coated cells are isolated by magnetic selection and deep sequencing of their barcodes is used to identify the displayed antigens. To benchmark the performance of REAP, we screened 77 patients with autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED). REAP sensitively and specifically detected known autoantibody reactivities in APECED in addition to numerous previously unidentified reactivities. We further screened 106 patients with systemic lupus erythematosus (SLE) and identified novel autoantibody reactivities against a diverse set of antigens including growth factors, extracellular matrix components, cytokines, and immunomodulatory proteins. Several of these responses were associated with disease severity and specific clinical manifestations of SLE and exerted potent functional effects on cell signaling ex vivo. These findings demonstrate the utility of REAP to atlas the expansive landscape of exoproteome-targeting autoantibodies and their impacts on patient health outcomes.

immunology↗

bioRxiv2021

DECODE: A Deep-learning Framework for Condensing Enhancers and Refining Boundaries with Large-scale Functional Assays

SummaryMapping distal regulatory elements, such as enhancers, is the cornerstone for investigating genome evolution, understanding critical biological functions, and ultimately elucidating how genetic variations may influence diseases. Previous enhancer prediction methods have used either unsupervised approaches or supervised methods with limited training data. Moreover, past approaches have operationalized enhancer discovery as a binary classification problem without accurate enhancer boundary detection, producing low-resolution annotations with redundant regions and reducing the statistical power for downstream analyses (e.g., causal variant mapping and functional validations). Here, we addressed these challenges via a two-step model called DECODE. First, we employed direct enhancer activity readouts from novel functional characterization assays, such as STARR-seq, to train a deep neural network classifier for accurate cell-type-specific enhancer prediction. Second, to improve the annotation resolution ([~]500 bp), we implemented a weakly-supervised object detection framework for enhancer localization with precise boundary detection (at 10 bp resolution) using gradient-weighted class activation mapping. ResultsOur DECODE binary classifier outperformed the state-of-the-art enhancer prediction methods by 24% in transgenic mouse validation. Further, DECODE object detection can condense enhancer annotations to only 12.6% of the original size, while still reporting higher conservation scores and genome-wide association study variant enrichments. Overall, DECODE improves the efficiency of regulatory element mapping with graphic processing units for deep-learning applications and is a powerful tool for enhancer prediction and boundary localization. Contactpi@gersteinlab.org

bioinformatics↗

bioRxiv2020

MiMeNet: Exploring Microbiome-Metabolome Relationships using Neural Networks

The advance in microbiome and metabolome studies has generated rich omics data revealing the involvement of the microbial community in host disease pathogenesis through interactions with their host at a metabolic level. However, the computational tools to uncover these relationships are just emerging. Here, we present MiMeNet, a neural network framework for modeling microbe-metabolite relationships. Using ten iterations of 10-fold cross-validation on three paired microbiome-metabolome datasets, we show that MiMeNet more accurately predicts metabolite abundances (mean Spearman correlation coefficients increase from 0.108 to 0.309, 0.276 to 0.457, and -0.272 to 0.264) and identifies more well-predicted metabolites (increase in the number of well-predicted metabolites from 198 to 366, 104 to 143, and 4 to 29) compared to state-of-art linear models for individual metabolite predictions. Additionally, we demonstrate that MiMeNet can group microbes and metabolites with similar interaction patterns and functions to illuminate the underlying structure of the microbe-metabolite interaction network, which could potentially shed light on uncharacterized metabolites through "Guilt by Association". Our results demonstrated that MiMeNet is a powerful tool to provide insights into the causes of metabolic dysregulation in disease, facilitating future hypothesis generation at the interface of the microbiome and metabolomics.

bioinformatics↗

bioRxiv2020

A novel family of nonribosomal peptides modulate collective behavior in Pseudovibrio bacteria isolated from marine sponges

Collective behavior is a common feature of life. Although swarming motility and biofilms are opposed collective behaviors, both contribute to bacterial survival and host colonization. We have identified a link between motility/biofilms and a nonribosomal peptide synthetase-polyketide synthase gene cluster family (ppp) conserved in Pseudovibrio and Pseudomonas Proteobacteria known to interact with diverse eukaryotes. After developing reverse genetics for Pseudovibrio, we discovered two pseudovibriamide families, heptapeptides with a reversal in chain polarity via an ureido linkage 1-6 and related nonadepsipeptides 7-12. Imaging mass spectrometry showed that 1 was excreted whereas 7 was colony-associated. Deletion of pppA abolished production of 1-12 leading to reduced motility and increased biofilm production. pppD mutants that produced only 1-6 showed motility comparable to the wild-type and reduced biofilm formation, indicating that the excreted heptapeptides play a role in promoting motility. In contrast to lipopeptides widely known to affect swarming and biofilms, pseudovibriamides are not surfactants. Our results expand current knowledge on metabolites mediating bacterial collective behavior. Moreover, the establishment of reverse genetics will enable future exploration of the ecological and biotechnological potential of Pseudovibrio bacteria which have been proposed to contribute to marine sponge health. SignificanceBacteria contribute to health and disease of plants and animals. Specialized metabolites produced by bacteria are important in mediating their behavior and the colonization of their hosts. We have identified a conserved gene cluster family in Pseudovibrio and Pseudomonas bacteria known to colonize marine animals and terrestrial plants, respectively. Using Pseudovibrio as a model, we show the encoded metabolites, which we termed pseudovibriamides, promote motility and decrease biofilms. In contrast to lipopeptides widely known to affect motility/biofilms, pseudovibriamides are not surfactants, but instead are linear peptides with a reversal in chain polarity. The discovery of pseudovibriamides expands current knowledge of bacteria collective behavior. The establishment of reverse genetics will enable exploration of the ecological and biotechnological potential of Pseudovibrio bacteria. ClassificationBiological Sciences, Microbiology

microbiology↗

bioRxiv2020

Characterization and description of Faecalibacterium butyricigenerans sp. nov. and F. longum sp. nov., isolated from human faeces

Exploiting a pure culture strategy to investigate the composition of human gut microbiota, two novel anaerobes, designated strains AF52-21T and CM04-06T, were isolated from faeces of two healthy Chinese donors and characterized using a polyphasic approach. The two strains were Gram-stain-negative, non-motile, and rod-shaped. Both strains grew optimally at 37{degrees}C and pH 7.0. Phylogenetic analysis based on 16S rRNA gene sequences revealed that the two strain clustered with species of the genus Faecalibacterium and were most closely related to Faecalibacterium prausnitzii ATCC 27768T with sequence similarity of 97.18% and 96.87%, respectively. The two isolates shared a 16S rRNA gene sequence identity of 98.69%. Draft genome sequencing was performed for strains AF52-21T and CM04-06T, generating genome sizes of 2.85 Mbp and 3.01 Mbp. The calculated average nucleotide identity values between the genomes of the strains AF52-21T and CM04-06T compared to Faecalibacterium prausnitzii ATCC 27768T were 83.20% and 82.54%, respectively, and 90.09% when comparing AF52-21T and CM04-06T. Both values were below the previously proposed species threshold (95%), supporting their recognition as novel species in the genus Faecalibacterium. The genomic DNA G+C contents of strain AF52-21T and CM04-06T calculated from genome sequences were 57.77 mol% and 57.51 mol%, respectively. Based on the phenotypic, chemotaxonomic and phylogenetic characteristics, we conclude that both strains represent two new Faecalibacterium species, for which the names Faecalibacterium butyricigenerans sp. nov. (type strain AF52-21T = CGMCC 1.5206T = DSM 103434T) and Faecalibacterium longum sp. nov. (type strain CM04-06T = CGMCC 1.5208T = DSM 103432T) are proposed.

microbiology↗

bioRxiv2020

Cardiac Gq receptors and calcineurin activation are not required for the hypertrophic response to mechanical left ventricular pressure overload

RationaleGq-coupled receptors are thought to play a critical role in the induction of left ventricular hypertrophy (LVH) secondary to pressure overload, although mechano-sensitive channel activation by a variety of mechanisms has also been proposed, and the relative importance of calcineurin- and calmodulin kinase II (CaMKII)-dependent hypertrophic pathways remains controversial. ObjectiveTo determine the mechanisms regulating the induction of LVH in response to mechanical pressure overload. Methods and ResultsTransgenic mice with cardiac-targeted inhibition of Gq-coupled receptors (GqI mice) and their non-transgenic littermates (NTL) were subjected to neurohumoral stimulation (continuous, subcutaneous angiotensin II (AngII) infusion for 14 days) or mechanical pressure overload (transverse aortic arch constriction (TAC) for 21 days) to induce LVH. Candidate signalling pathway activation was examined. As expected, LVH observed in NTL mice with AngII infusion was attenuated in heterozygous (GqI+/-) mice and absent in homozygous (GqI-/-) mice. In contrast, LVH due to TAC was unaltered by either heterozygous or homozygous Gq inhibition. Gene expression of atrial natriuretic peptide (ANP), B-type natriuretic peptide (BNP) and -skeletal actin (-SA) was increased 48 hours after AngII infusion or TAC in NTL mice; in GqI mice, the increases in ANP, BNP and -SA in response to AngII were completely absent, as expected, but all three increased after TAC. Increased nuclear translocation of nuclear factor of activated T-cells c4 (NFATc4), indicating calcineurin pathway activation, occurred in NTL mice with AngII infusion but not TAC, and was prevented in GqI mice infused with AngII. Nuclear and cytoplasmic CaMKII{delta} levels increased in both NTL and GqI mice after TAC but not AngII infusion, with increased cytoplasmic phospho- and total histone deacetylase 4 (HDAC4) and increased nuclear myocyte enhancer factor 2 (MEF2) levels. ConclusionCardiac Gq receptors and calcineurin activation are required for neurohumorally mediated LVH but are not required for LVH induced by mechanical pressure overload (TAC); the latter is mediated by activation of the CaMKII-HDAC4-MEF2 pathway.

molecular biology↗

bioRxiv2020

Mitochondrial fusion mediated by mitofusin 1 regulates macrophage mycobactericidal activity by enhancing autophagy

Mitochondria as a highly dynamic organelle continuously changes morphology and position during its life cycle. Mitochondrial dynamics including fission and fusion play a critical role in maintaining functional mitochondria for ATP production, which is directly linked to host defense against Mtb infection. However, how macrophages regulate mitochondrial dynamics during Mycobacterium tuberculosis (Mtb) infection remains elusive. In this study, we found that Mtb infection induced mitochondrial fusion through enhancing the expression of mitofusin 1 (MFN1), which resulted in increased ATP production. Silencing MFN1 inhibited mitochondrial fusion and subsequently reduced ATP production, which, in turn, severely impaired macrophage mycobactericidal activity by inhibiting autophagy. Impairment of mycobactericidal activity and autophagy was replicated using oligomycin, an inhibitor of ATP synthase. In summary, our study revealed MFN1-mediated mitochondrial fusion is essential for macrophage mycobactericidal activity through the regulation of ATP dependent autophagy. MFN1-mediated metabolism pathway might be targets for development of host direct therapy (HDT) strategy against TB. ImportanceHow mitochondrial dynamic is regulated in attempt to fight against Mtb remains elusive. Our study revealed that the fission/fusion dynamics of mitochondria during Mtb infection is regulated by MFN1, through which mitochondrial respiration and autophagy activity are affected. Our findings suggested that intervention of energy metabolism by targeting MFN1 might be a strategy against Mtb infection.

immunology↗

bioRxiv2020

Taxonomic description and genome sequence of Christensenella intestinihominis sp. nov., a novel cholesterol-lowering bacterium isolated from human gut

A Gram-staining-negative, non-spore-forming, short, straight rod, non-motile, obligate anaerobic bacterial strain, AF73-05CM02T, was isolated from faecal sample of thirty-years old healthy male. Colonies were approximately 0.2mm in diameter, beige and circular after 4 days of incubation on PYG agar under anaerobic conditions at 37{degrees}C. Strain AF73-05CM02T grew in a temperature range between 30-42{degrees}C and pH range from 6.0 to 8.5, with optimum growth at 37-42{degrees}C and pH 7.0. Base on the 16S rRNA gene sequence analysis, strain AF73-05CM02T belongs to the genus Christensenella and showed the highest level of sequence similarity with Christensenella minuta YIT 12065T. The predominant fatty acids of strain AF73-05CM02T were C10:0 (7.5%), iso-C11:0 (5.6%), C12:0 (7.2%), C14:0 (46.6%), iso-C15:0 (7.4%), C16:0 (9.7%) and C18:1 {omega} 9 c (6.9%). Acetic acid, formic acid, butyric acid and lactic acid were end products of glucose fermentation. The isolate was negative for catalase, indole production and hydrolysis of gelatin. Genomic relatedness analyses based on average nucleotide identity indicated strain AF73-05CM02T significantly differed from other species of the genus Christensenella, showing ANI values under 82.89% with the phylogenetically closest species. The G+C content of the genomic DNA was 52.07 mol% from genome sequence, which differs from that of Christensenella minuta. Several physiological biochemical and genotypic properties differentiated the novel bacterial strain from the related species, indicating that the isolate represent a new species of the genus Christensenella for which the name Christensenella intestinihominis sp. nov. is proposed, with strain AF73-05CM02T (=CGMCC 1.5207T =DSM 103477T) being the type strain. The follow study approached the function of cholesterol-lowering for strains AF73-05CM02T and DSM 22067T revealed the two strains exhibit capacity for removing cholesterol with efficiency of 36.6% and 54.3%. Exopolysaccharide production of two strains were 234 and 271 mg/L, respectively.

microbiology↗

bioRxiv2020

Essential gene analysis in Acinetobacter baumannii by high-density transposon mutagenesis and CRISPR interference

Acinetobacter baumannii is a poorly understood bacterium capable of life-threatening infections in hospitals. Few antibiotics remain effective against this highly resistant pathogen. Developing rationally-designed antimicrobials that can target A. baumannii requires improved knowledge of the proteins that carry out essential processes allowing growth of the organism. Unfortunately, studying essential genes has been challenging using traditional techniques, which usually require time-consuming recombination-based genetic manipulations. Here, we performed saturating mutagenesis with dual transposon systems to identify essential genes in A. baumannii and we developed a CRISPR-interference (CRISPRi) system for facile analysis of these genes. We show that the CRISPRi system enables efficient transcriptional silencing in A. baumannii. Using these tools, we confirmed the essentiality of the novel cell division protein AdvA and discovered a previously uncharacterized AraC-family transcription factor (ACX60_RS03245) that is necessary for growth. In addition, we show that capsule biosynthesis is a conditionally essential process, with mutations in late-acting steps causing toxicity in strain ATCC 17978 that can be bypassed by blocking early-acting steps or activating the BfmRS stress response. These results open new avenues for analysis of essential pathways in A. baumannii. ImportanceNew approaches are urgently needed to control A. baumannii, one of the most drug resistant pathogens known. To facilitate the development of novel targets that allow inhibition of the pathogen, we performed a large-scale identification of genes whose products the bacterium needs for growth. We also developed a CRISPR-based gene knockdown tool that operates efficiently in A. baumannii, allowing rapid analysis of these essential genes. We used these methods to define multiple processes vital to the bacterium, including a previously uncharacterized gene-regulatory factor and export of a protective polymeric capsule. These tools will enhance our ability to investigate processes critical for the essential biology of this challenging hospital-acquired pathogen.

microbiology↗

bioRxiv2020

Immunologically distinct responses occur in the CNS of COVID-19 patients

One third of COVID-19 patients develop significant neurological symptoms, yet SARS-CoV-2 is rarely detected in central nervous system (CNS) tissue, suggesting a potential role for parainfectious processes, including neuroimmune responses. We therefore examined immune parameters in cerebrospinal fluid (CSF) and blood samples from a cohort of patients with COVID-19 and significant neurological complications. We found divergent immunological responses in the CNS compartment, including increased levels of IL-12 and IL-12-associated innate and adaptive immune cell activation. Moreover, we found increased proportions of B cells in the CSF relative to the periphery and evidence of clonal expansion of CSF B cells, suggesting a divergent intrathecal humoral response to SARS-CoV-2. Indeed, all COVID-19 cases examined had anti-SARS-CoV-2 IgG antibodies in the CSF whose target epitopes diverged from serum antibodies. We directly examined whether CSF resident antibodies target self-antigens and found a significant burden of CNS autoimmunity, with the CSF from most patients recognizing neural self-antigens. Finally, we produced a panel of monoclonal antibodies from patients CSF and show that these target both anti-viral and anti-neural antigens--including one mAb specific for the spike protein that also recognizes neural tissue. This exploratory immune survey reveals evidence of a compartmentalized and self-reactive immune response in the CNS meriting a more systematic evaluation of neurologically impaired COVID-19 patients. One Sentence SummaryA subset of COVID-19 patients with neurologic impairment show cerebrospinal fluid-specific immune alterations that point to both neuroinvasion and anti-neural autoimmunity as potential causes of impairment.

immunology↗

bioRxiv2020

A single intranasal dose of chimpanzee adenovirus-vectored vaccine confers sterilizing immunity against SARS-CoV-2 infection

The Coronavirus Disease 2019 pandemic has made deployment of an effective vaccine a global health priority. We evaluated the protective activity of a chimpanzee adenovirus-vectored vaccine encoding a pre-fusion stabilized spike protein (ChAd-SARS-CoV-2-S) in challenge studies with Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and mice expressing the human angiotensin-converting enzyme 2 receptor. Intramuscular dosing of ChAd-SARS-CoV-2-S induces robust systemic humoral and cell-mediated immune responses and protects against lung infection, inflammation, and pathology but does not confer sterilizing immunity, as evidenced by detection of viral RNA and induction of anti-nucleoprotein antibodies after SARS-CoV-2 challenge. In contrast, a single intranasal dose of ChAd-SARS-CoV-2-S induces high levels of systemic and mucosal IgA and T cell responses, completely prevents SARS-CoV-2 infection in the upper and lower respiratory tracts, and likely confers sterilizing immunity in most animals. Intranasal administration of ChAd-SARS-CoV-2-S is a candidate for preventing SARS-CoV-2 infection and transmission, and curtailing pandemic spread.

microbiology↗

bioRxiv2020

Neuroinvasive potential of SARS-CoV-2 revealed in a human brain organoid model

Although COVID-19 is considered to be primarily a respiratory disease, SARS-CoV-2 affects multiple organ systems including the central nervous system (CNS). Yet, there is no consensus whether the virus can infect the brain, or what the consequences of CNS infection are. Here, we used three independent approaches to probe the capacity of SARS-CoV-2 to infect the brain. First, using human brain organoids, we observed clear evidence of infection with accompanying metabolic changes in the infected and neighboring neurons. However, no evidence for the type I interferon responses was detected. We demonstrate that neuronal infection can be prevented either by blocking ACE2 with antibodies or by administering cerebrospinal fluid from a COVID-19 patient. Second, using mice overexpressing human ACE2, we demonstrate in vivo that SARS-CoV-2 neuroinvasion, but not respiratory infection, is associated with mortality. Finally, in brain autopsy from patients who died of COVID-19, we detect SARS-CoV-2 in the cortical neurons, and note pathologic features associated with infection with minimal immune cell infiltrates. These results provide evidence for the neuroinvasive capacity of SARS-CoV2, and an unexpected consequence of direct infection of neurons by SARS-CoV-2.

microbiology↗

bioRxiv2020

Polyclonal epitope cartography reveals the temporal dynamics and diversity of human antibody responses to H5N1 vaccination

Novel influenza A virus (IAV) strains elicit recall immune responses to conserved epitopes, making them favorable antigenic choices for universal influenza virus vaccines. Evaluating these immunogens requires a thorough understanding of the antigenic sites targeted by the polyclonal antibody (pAb) response, which single particle electron microscopy (EM) can sensitively detect. Here, we employed EM polyclonal epitope mapping (EMPEM) to extensively characterize the pAb response to hemagglutinin (HA) after H5N1 immunization in humans. Cross-reactive pAbs originating from memory B cells immediately bound the stem of HA and persisted for over a year post vaccination. In contrast, de novo pAb responses to multiple sites on the head of HA, which targeted previously determined key neutralizing sites on H5 HA, expanded after the second immunization and waned quickly. Thus, EMPEM provides a robust tool for comprehensively tracking the specificity and durability of immune responses elicited by novel universal influenza vaccine candidates.

immunology↗

bioRxiv2020

Using Conditional Generative Adversarial Networks to Boost the Performance of Machine Learning in Microbiome Datasets

The microbiome of the human body has been shown to have profound effects on physiological regulation and disease pathogenesis. However, association analysis based on statistical modeling of microbiome data has continued to be a challenge due to inherent noise, complexity of the data, and high cost of collecting large number of samples. To address this challenge, we employed a deep learning framework to construct a data-driven simulation of microbiome data using a conditional generative adversarial network. Conditional generative adversarial networks train two models against each other while leveraging side information learn from a given dataset to compute larger simulated datasets that are representative of the original dataset. In our study, we used a cohorts of patients with inflammatory bowel disease to show that not only can the generative adversarial network generate samples representative of the original data based on multiple diversity metrics, but also that training machine learning models on the synthetic samples can improve disease prediction through data augmentation. In addition, we also show that the synthetic samples generated by this cohort can boost disease prediction of a different external cohort.

bioinformatics↗

bioRxiv2020

Meta-Signer: Metagenomic Signature Identifier based on Rank Aggregation of Features

BackgroundThe advance of metagenomic studies provides the opportunity to identify microbial taxa that are associated to human diseases. Multiple methods exist for the association analysis. However, the results could be inconsistent, presenting challenges in interpreting the host-microbiome interactions. To address this issue, we introduce Meta-Signer, a novel Metagenomic Signature Identifier tool based on rank aggregation of features identified from multiple machine learning models including Random Forest, Support Vector Machines, LASSO, Multi-Layer Perceptron Neural Networks, and our recently developed Convolutional Neural Network framework (PopPhy-CNN). Meta-Signer generates ranked taxa lists by training individual machine learning models over multiple training partitions and aggregates them into a single ranked list by an optimization procedure to represent the most informative and robust microbial features. Meta-Signer can rank taxa using two input forms of the data: the relative abundances of the original taxa and taxa from the populated taxonomic trees generated from the original taxa. The latter form allows the evaluation of the association of microbial features at different taxonomic levels to the disease, which is attributed to our novel model of PopPhy-CNN. ResultsWe evaluate Mega-Signer on five different human gut-microbiome datasets. We demonstrate that the features derived from Meta-Signer were more informative compared to those obtained from other available feature ranking methods. The highly ranked features are strongly supported by published literature. ConclusionMeta-Signer is capable of deriving a robust set of microbial features at multiple taxonomic levels for the prediction of host phenotype. Meta-Signer is user-friendly and customizable, allowing users to explore their datasets quickly and efficiently.

bioinformatics↗

bioRxiv2020

Antibiotic hypersensitivity signatures identify targets for attack in the Acinetobacter baumannii cell envelope

Acinetobacter baumannii is an opportunistic pathogen that is a critical, high-priority target for new antibiotic development. Clearing of A. baumannii requires relatively high doses of antibiotics across the spectrum, primarily due to its protective cell envelope. Many of the proteins that support envelope integrity and modulate drug action are uncharacterized, largely because there is an absence of orthologs for several proteins that perform essential envelope-associated processes, impeding progress on this front. To identify targets that can synergize with current antibiotics, we performed an exhaustive analysis of A. baumannii mutants causing hypersensitivity to a multitude of antibiotic treatments. By examining mutants with antibiotic hypersensitivity profiles that parallel mutations in proteins of known function, we show that the function of poorly annotated proteins can be predicted and used to identify candidate missing link proteins in essential A. baumannii processes. Using this strategy, we uncovered multiple uncharacterized proteins with critical roles in cell division or cell elongation, and revealed that a predicted cell wall D,D-endopeptidase has an unappreciated function in lipooligosaccharide synthesis. Moreover, we provide a genetic strategy that uses hypersensitivity signatures to predict drug synergies, allowing the identification of {beta}-lactams that work cooperatively based on the cell wall assembly machineries that they preferentially target. These data reveal multiple pathways critical for envelope growth in A. baumannii that can be targeted in combination strategies for attacking the pathogen.

microbiology↗