Biology subjects

Tian, L.

Publications and source records attributed to Tian, L..

178 records · Page 10Linked to original sources

bioRxiv2019

CSF sTREM2 correlates with CSF tau in advancing Parkinson’s disease

Parkinsons disease (PD) is the second most common neurodegenerative disease after Alzheimers disease (AD) and affects 1% of the population above 60 years old. Although PD commonly manifests with motor symptoms, a majority of patients with PD subsequently develop cognitive impairment which often progresses to dementia, a major cause of morbidity and disability. PD is characterized by -synuclein accumulation that frequently associates with amyloid beta (A{beta}) and tau fibrils, the hallmarks of AD neuropathologic changes; this co-occurrence suggests that onset of cognitive decline in PD may be associated with appearance of pathologic A{beta} and/or tau. Recent studies have highlighted the appearance of the soluble form of the Triggering Receptor Expressed on Myeloid cells 2 (sTREM2) receptor in CSF during development of AD. Given the known association of microglial activation with advancing PD, we investigated whether CSF and/or plasma sTREM2 increased with progression to PD dementia. We examined 165 participants consisting of 17 cognitively normal elderly, 45 PD patients with no cognitive impairment, 86 with mild cognitive impairment, and 17 with dementia. Stratification of subjects by CSF A{beta} and tau levels revealed that CSF sTREM2 concentrations were elevated in PD subgroups with abnormal tau, but not A{beta}, CSF concentration. These findings indicate that CSF sTREM2 could serve as a surrogate immune biomarker of neuronal injury in PD that is associated with cognitive decline.\n\nOne sentence summaryCSF sTREM2 correlates with CSF tau in PD

neuroscience↗

bioRxiv2019

Deep spectral learning for label-free optical imaging oximetry with uncertainty quantification

Measurement of blood oxygen saturation (sO2) by optical imaging oximetry provides invaluable insight into local tissue functions and metabolism. Despite different embodiments and modalities, all label-free optical imaging oximetry utilize the same principle of sO2-dependent spectral contrast from hemoglobin. Traditional approaches for quantifying sO2 often rely on analytical models that are fitted by the spectral measurements. These approaches in practice suffer from uncertainties due to biological variability, tissue geometry, light scattering, systemic spectral bias, and variations in experimental conditions. Here, we propose a new data-driven approach, termed deep spectral learning (DSL) for oximetry to be highly robust to experimental variations, and more importantly to provide uncertainty quantification for each sO2 prediction. To demonstrate the robustness and generalizability of DSL, we analyze data from two visible light optical coherence tomography (vis-OCT) setups across two separate in vivo experiments in rat retina. Predictions made by DSL are highly adaptive to experimental variabilities as well as the depth-dependent backscattering spectra. Two neural-network-based models are tested and compared with the traditional least-squares fitting (LSF) method. The DSL-predicted sO2 shows significantly lower mean-square errors than the LSF. For the first time, we have demonstrated en face maps of retinal oximetry along with pixel-wise confidence assessment. Our DSL overcomes several limitations in the traditional approaches and provides a more flexible, robust, and reliable deep learning approach for in vivo non-invasive label-free optical oximetry.

bioengineering↗

bioRxiv2019

Germline-encoded TCR-MHC contacts promote TCR V gene bias in umbilical cord blood T cell repertoire

T cells recognize antigens as peptides bound to major histocompatibility complex (MHC) proteins through T cell receptors (TCRs) on their surface. To recognize a wide range of pathogens, each individual possesses a substantial number of TCRs with an extremely high degree of variability. It remains controversial whether germline-encoded TCR repertoire is shaped by MHC polymorphism and, if so, what is the preference between MHC genetic variants and TCR V gene compatibility. To investigate the \"net\" genetic association between MHC variations and TRBV genes, we applied quantitative trait locus (QTL) mapping to test the associations between MHC polymorphism and TCR {beta} chain V (TRBV) genes usage using umbilical cord blood (UCB) samples of 201 Chinese newborns. We found TRBV gene and MHC loci that are predisposed to interact with one another differ from previous conclusions. The majority of MHC amino acid residues associated with the TRBV gene usage show spatial proximities in known structures of TCR-pMHC complexes. These results show for the first time that MHC variants bias TRBV gene usage in UCB of Chinese ancestry and indicate that germline-encoded contacts influence TCR-MHC interactions in intact T cell repertoires.

immunology↗

bioRxiv2019

All-Assay-Max2 pQSAR: Activity predictions as accurate as 4-concentration IC50s for 8,558 Novartis assays

Profile-QSAR (pQSAR) is a massively multi-task, 2-step machine learning method with unprecedented scope, accuracy and applicability domain. In step one, a \"profile\" of conventional single-assay random forest regression (RFR) models are trained on a very large number of biochemical and cellular pIC50 assays using Morgan 2 sub-structural fingerprints as compound descriptors. In step two, a panel of PLS models are built using the profile of pIC50 predictions from those RFR models as compound descriptors. Hence the name. Previously described for a panel of 728 biochemical and cellular kinase assays, we have now built an enormous pQSAR from 11,805 diverse Novartis IC50 and EC50 assays. This large number of assays, and hence of compound descriptors for PLS, dictated reducing the profile by only including RFR models whose predictions correlate with the assay being modeled. The RFR and pQSAR models were evaluated with our \"realistically novel\" held-out test set whose median average similarity to the nearest training set member across the 11,805 assays was only 0.34, thus testing a realistically large applicability domain. For the 11,805 single-assay RFR models, the median correlation of prediction with experiment was only R2 ext=0.05, virtually random, and only 8% of the models achieved our standard success threshold of R2 ext=0.30. For pQSAR, the median correlation was R2 ext=0.53, comparable to 4-concentration experimental IC50s, and 72% of the models met our R2 ext>0.30 standard, totaling 8558 successful models. The successful models included assays from all of the 51 annotated target sub-classes, as well as 4196 phenotypic assays, indicating that pQSAR can be applied to virtually any disease area. Every month, all models are updated to include new measurements, and predictions are made for 5.5 million Novartis compounds, totaling 50 billion predictions. Common uses have included virtual screening, selectivity design, toxicity and promiscuity prediction, mechanism-of-action prediction, and others.

bioinformatics↗

bioRxiv2019

SynQuant: An Automatic Tool to Quantify Synapses from Microscopy Images

MotivationSynapses are essential to neural signal transmission. Therefore, quantification of synapses and related neurites from images is vital to gain insights into the underlying pathways of brain functionality and diseases. Despite the wide availability of synapse imaging data, several issues prevent satisfactory quantification of these structures by current tools. First, the antibodies used for labeling synapses are not perfectly specific to synapses. These antibodies may exist in neurites or other cell compartments. Second, the brightness for different neurites and synapses is heterogeneous due to the variation of antibody concentration and synapse-intrinsic differences. Third, images often have low signal to noise ratio (SNR) due to constraints of experiments and availability of sensitive antibodies. The combination of these issues makes the detection of synapses challenging and necessitates developing a new tool to accurately and reliably quantify synapses. ResultsWe present an automatic probability-principled synapse detection algorithm and integrate it into our synapse quantification tool SynQuant. Derived from the theory of order statistics, our method controls the false discovery rate and improves the power of detecting synapses. Through extensive experiments on both synthetic and real images in the presence of severe antibody diffusion, high heterogeneity, and large noise, our method was demonstrated to outperform peer specialized synapse detection tools as well as generic spot detection methods by a large margin. Finally, we show SynQuant reliably uncovers statistically significant differences between disease and control conditions in a neuron-astrocyte co-culture based model of Down Syndrome. AvailabilityThe Java source code, Fiji plug-in, and test data are available at https://github.com/yu-lab-vt/SynQuant. Contactyug@vt.edu

neuroscience↗

bioRxiv2019

The vagino-cervical microbiome as a woman's life history

The gut microbiome has been the center of attention for human commensal microbiome studies. The vaginal microbiome is also densely populated with bacteria, viruses and fungi, and the presence of microorganisms beyond the cervix is increasingly reported in non-infectious conditions1-3. Due to the over 90% of human sequences in female reproductive tract samples3,4, metagenomic information has been very limited. 16S rRNA gene amplicon sequencing studies have identified community types in the vaginal microbiota, and observed its dynamic changes due to menstrual cycles and sexual behaviors in small cohorts5,6. Here we perform metagenomic shotgun sequencing on cervical samples from 516 women of reproductive age (more than 10-fold of the Human Microbiome Project (HMP)4), and dissect major factors, especially pregnancy and delivery histories and contraception methods on the microbiome composition. Features of other body sites, such as mood fluctuations and facial speckles could potentially be deduced from the vagino-cervical microbiome. Our results offer an unprecedented glimpse into the microbiota in the female reproductive tract and imply disease susceptibilities that may be relieved by behavioural changes.

microbiology↗

bioRxiv2019

Spatial organization of the transcriptional regulatory network of Saccharomyces cerevisiae

Transcriptional regulatory network (TRN) is a directed complex network composed of all regulatory interactions between transcription factors and corresponding target genes. Recently, the three-dimensional (3D) genomics studies have shown that the 3D structure of the genome makes a difference to the regulation of gene transcription, which provides us with a novel perspective. In this study, we constructed the TRN of the budding yeast Saccharomyces cerevisiae and placed it in the context of 3D genome model. We analyzed the spatial organization of the yeast TRN on four levels: global feature, central nodes, hierarchical structure and network motifs. Our results suggested that the TRN of S. cerevisiae presents an optimized structure in space to adapt to functional requirement.

systems biology↗

bioRxiv2018

Introduction and validation of a new semi-automated method to determine sympathetic fiber density in target tissues.

In recent years, the role of sympathetic nervous fibers in chronic inflammation has become increasingly evident. At the onset of inflammation, sympathetic activity is increased in the affected tissue. However, sympathetic fibers are largely absent from chronically inflamed tissue. Apparently, there is a very dynamic relationship between sympathetic innervation and the immune system in areas of inflammation, and hence a rapid and easy method for quantification of nerve fiber density of target organs is of great value to answer potential research questions. Currently, nervous fiber densities are either determined by tedious manual counting, which is not suitable for high throughput approaches, or by expensive automated processes relying on specialized software and high-end microscopy equipment. Usually, tyrosine hydroxylase (TH) is used as the marker for sympathetic fibers. In order to overcome the current quantification bottleneck with a cost-efficient alternative, an automated process was established and compared to the classic manual approach of counting TH-positive sympathetic fibers. Since TH is not exclusively expressed on sympathetic fibers, but also in a number of catecholamine-producing cells, a prerequisite for automated determination of fiber densities is to reliably distinct between cells and fibers. Therefore, an additional staining using peripherin exclusively expressed in nervous fibers as a secondary marker was established. Using this novel approach, we studied the spleens from a syndecan-3 knockout (SDC3KO) mouse line, and demonstrated equal results on SNS fiber density for both manual and automated counts (Manual counts: wt: 22.57 +/- 11.72 fibers per mm2; ko: 31.95 +/- 18.85 fibers per mm2; p = 0.0498; Automated counts: wt: 31.6 +/- 18.98 fibers per mm2; ko: 45.49 +/- 19.65 fibers per mm2; p = 0.01868). In conclusion, this new and simple method can be used as a high-throughput approach to reliably and quickly quantify SNS nerve fiber density in target tissues.

neuroscience↗

bioRxiv2018

scRNA-seq mixology: towards better benchmarking of single cell RNA-seq protocols and analysis methods

Single cell RNA sequencing (scRNA-seq) technology has undergone rapid development in recent years, bringing with it new challenges in data processing and analysis. This has led to an explosion of tailored analysis methods for scRNA-seq to address various biological questions. However, the current lack of gold-standard benchmarking datasets makes it difficult for researchers to evaluate the performance of the many methods. Here, we designed and carried out a realistic benchmark experiment that included mixtures of single cells or pseudo-cells created by sampling admixtures of cells or RNA from 3 distinct cancer cell lines. Altogether we generated 10 datasets using a combination of droplet and plate-based scRNA-seq protocols, with varying data quality, population heterogeneity and noise levels. Using these benchmark datasets, we compared different protocols, evaluated the spike-in standard and multiple data analysis methods for tasks ranging from normalization and imputation, to clustering, trajectory analysis and data integration. Evaluation of methods across multiple datasets revealed some that performed well in general and others that suited specific situations. Our dataset and analysis provide a comprehensive comparison framework for benchmarking most popular scRNA-seq analysis tasks.

bioinformatics↗

bioRxiv2018

Changes in pregnancy-related serum biomarkers early in gestation are associated with later development of preeclampsia

BackgroundPlacental protein expression plays a crucial biological role during normal and complicated pregnancies. We hypothesized that: (1) circulating pregnancy-associated, placenta-related protein levels throughout gestation reflect the uncomplicated, full-term temporal progression of human gestation, and effectively estimates gestational ages (GAs); (2) pregnancies with underlying placental pathology, such as preeclampsia (PE), are associated with disruptions in this GA estimation in early gestation; (3) malfunctions of this GA estimation can be employed to identify impending PE. In addition, to explore the underlying biology and PE etiology, we set to compare protein gestational patterns of human and mouse, using pregnant heme oxygenase-1 (HO-1) heterozygote (Het) mice, a mouse model reflecting PE-like symptoms.\n\nMethodsSerum levels of circulating placenta-related proteins - leptin (LEP), chorionic somatomammotropin hormone like 1 (CSHL1), elabela (ELA), activin A, soluble fms-like tyrosine kinase 1 (sFlt-1), and placental growth factor (PlGF)- were quantified by ELISA in blood serially collected throughout human pregnancies (20 normal subjects with 66 samples, and 20 PE subjects with 61 samples). Linear multivariate analysis of the targeted serological protein levels was performed to estimate the normal GA. Logarithmic transformed mean-squared errors of GA estimations were used to identify impending PE. Then the human gestational protein patterns were compared to those in the pregnant HO-1 mice.\n\nResultsAn elastic net (EN)-based gestational dating model was developed (R2 = 0.76) and validated (R2 = 0.61) using the serum levels of the 6 proteins at various GAs from women with normal uncomplicated pregnancies (n = 10 for training and n = 6 for validation). In pregnancies complicated by PE (n = 14), the EN model was not (R2 = -0.17) associated with GA at sampling in PE. Statistically significant deviations from the normal GA EN model estimations were observed in PE-associated pregnancies between GAs of 16-30 weeks (P = 0.01). The EN model developed with 5 proteins (ELA excluded due to the lack of robustness of the mouse ELA essay) performed similarly on normal human (R2 = 0.68) and WT mouse (R2 = 0.85) pregnancies. Disruptions of this model were observed in both human PE-associated (human: R2 = 0.27) and mouse HO-1 Het (mouse: R2 = 0.30) pregnancies. LEP out performed sFlt-1 and PlGF in differentiating impending PE at early human and late mouse gestations.\n\nConclusionsAs revealed in both human and mouse GA EN analyses, temporal serological placenta-related protein patterns are tightly regulated throughout normal human pregnancies and can be significantly disrupted in pathologic PE states. LEP changes earlier during gestation than the well-established late GA PE biomarkers (sFlt-1 and PlGF). Our HO-1 Het mouse analysis provides direct evidence of the causative action of HO-1 deficiency in LEP upregulation in a PE-like murine model. Therefore, longitudinal analyses of pregnancy-related protein patterns in sera, may not only help in the exploration of underlying PE pathophysiology but also provide better clinical utility in PE assessment.

molecular biology↗

bioRxiv2018

Dopamine neurons projecting to medial shell of the nucleus accumbens drive heroin reinforcement

The dopamine (DA) hypothesis posits the increase of mesolimbic dopamine levels as a defining commonality of addictive drugs, initially causing reinforcement, eventually leading to compulsive consumption. While much experimental evidence from psychostimulants supports this hypothesis, it has been challenged for opioid reinforcement. Here, we use genetically encoded DA and calcium indicators as well as cFos to reveal that heroin activates DA neurons located in the medial part of the VTA, preferentially projecting to the medial shell of the nucleus accumbens (NAc). Chemogenetic and optogenetic manipulations of VTA DA or GABA neurons establish a causal link to heroin reinforcement. Inhibition of DA neurons blocked heroin self-administration, while heroin inhibited optogenetic self-stimulation of DA neurons. Likewise, heroin occluded the self-inhibition of VTA GABA neurons. Together, these experiments support a model of disinhibition of a subset of VTA DA neurons in opioid reinforcement.

neuroscience↗

bioRxiv2018

A large-scale whole-genome sequencing analysis reveals highly specific genome editing by both Cas9 and Cpf1 nucleases in rice

Targeting specificity has been an essential issue for applying genome editing systems in functional genomics, precise medicine and plant breeding. Understanding the scope of off-target mutations in Cas9 or Cpf1-edited crops is critical for research and regulation. In plants, only limited studies had used whole-genome sequencing (WGS) to test off-target effects of Cas9. However, the cause of numerous discovered mutations is still controversial. Furthermore, WGS based off-target analysis of Cpf1 has not been reported in any higher organism to date. Here, we conducted a WGS analysis of 34 plants edited by Cas9 and 15 plants edited by Cpf1 in T0 and T1 generations along with 20 diverse control plants in rice, a major food crop with a genome size of ~380 Mb. The sequencing depth ranged from 45X to 105X with reads mapping rate above 96%. Our results clearly show that most mutations in edited plants were created by tissue culture process, which caused ~102 to 148 single nucleotide variations (SNVs) and ~32 to 83 insertions/deletions (indels) per plant. Among 12 Cas9 single guide RNAs (sgRNAs) and 3 Cpf1 CRISPR RNAs (crRNAs) assessed by WGS, only one Cas9 sgRNA resulted in off-target mutations in T0 lines at sites predicted by computer programs. Moreover, we cannot find evidence for bona fide off-target mutations due to continued expression of Cas9 or Cpf1 with guide RNAs in T1 generation. Taken together, our comprehensive and rigorous analysis of WGS big data across multiple sample types suggests both Cas9 and Cpf1 nucleases are very specific in generating targeted DNA modifications and off-targeting can be avoided by designing guide RNAs with high specificity.

molecular biology↗

bioRxiv2018

Identification of Variable and Joining germline genes and alleles for Rhesus macaque from B-cell receptor repertoires

The Rhesus macaque is a valuable preclinical animal model to estimate vaccine effectiveness, and is also important for understanding antibody maturation and B-cell repertoire evolution responding to vaccination; however, incomplete mapping of rhesus immunoglobulin germline genes hinders the research efforts. To address this deficiency, we sequenced B-cell receptor (BCR) repertoires of 75 India Rhesus macaques. Using a bioinformatic method that has been validated with BCR repertoire analysis of three human donors, we were able to infer rhesus Variable (V) and Joint(J) germline alleles, identifying a total of 122 V and 20 J germline alleles. Importantly, 91 V and 13 J alleles were novel, and 40 V and 13 J genes were found at a novel genome region that has not been previously recorded. The novelty of these newly identified alleles was supported by two observations. Firstly, 50 V and 5 J novel alleles were observed in whole genome sequencing data of 10 Rhesus macaques. Secondly, using alignment reference including the novel alleles, the mutation rate of rearranged repertoires was significant declined in 9 other irrelevant samples, and all our identified novel V and J alleles were 100% identity mapped by rearranged repertoire data. These newly identified novel alleles, along with previous reported alleles, provide an important reference for future investigations of rhesus immune repertoire evolution, in response to vaccination or infection. In addition, the method outlined in our study offered an example to future efforts in identifying novel immunoglobulin alleles.

immunology↗

bioRxiv2018

Aberrant calcium signaling in astrocytes inhibits neuronal excitability in a human Down syndrome stem cell model

Down syndrome (DS) is a devastating genetic disorder causing severe cognitive impairment. The staggering array of effects associated with an extra copy of human chromosome 21 (HSA21) complicates mechanistic understanding of DS pathophysiology. We developed an in vitro system to examine the interplay of neurons and astrocytes in a fully recapitulated HSA21 trisomy model differentiated from DS patient-derived induced pluripotent stem cells (iPSCs). By combining calcium imaging with genetic approaches, we utilized this system to investigate the functional defects of DS astroglia and their effects on neuronal excitability. We found that, compared with control isogenic astroglia, DS astroglia exhibited more-frequent spontaneous calcium fluctuations, which reduced the excitability of co-cultured neurons. DS astrocytes exerted this effect on both DS and healthy neurons. Neuronal activity could be rescued by abolishing astrocytic spontaneous calcium activity either chemically by blocking adenosine-mediated astrocyte-neuron signaling or genetically by knockdown of inositol triphosphate (IP3) receptors or S100{beta}, a calcium binding protein coded on HSA21. Our results suggest a novel mechanism by which DS alters the function of astrocytes, which subsequently disturbs neuronal excitability. Furthermore, our study establishes an all-optical neurophysiological platform for studying human neuron-astrocyte interactions associated with neurological disorders.\n\nSignificant statementDown syndrome (DS) is the most common genetic disorder caused by trisomy of chromosome 21 (HSA21). Problems with cognitive impairment, have not been properly addressed due to the inability to fully recapitulate HSA21, which is further confounded by the snapshot views of morphological changes of brain cells in isolation obtained from current studies. The brain develops neural networks consisting of neurons and glial cells that work together. To understand how DS affects the neural networks, we used DS patient-derived stem cells and calcium imaging to investigate functional defects of DS astrocytes and their effects on neuronal excitability. Our study has significant implication in understanding functional defects during brain development underlying DS.

neuroscience↗

bioRxiv2017

scPipe: a flexible data preprocessing pipeline for single-cell RNA-sequencing data

Single-cell RNA sequencing (scRNA-seq) technology allows researchers to profile the transcriptomes of thousands of cells simultaneously. Protocols that incorpo-rate both designed and random barcodes have greatly increased the throughput of scRNA-seq, but give rise to a more complex data structure. There is a need for new tools that can handle the various barcoding strategies used by different protocols and exploit this information for quality assessment at the sample-level and provide effective visualization of these results in preparation for higher-level analyses.\n\nTo this end, we developed scPipe, a R/Bioconductor package that integrates barcode demultiplexing, read alignment, UMI-aware gene-level quantification and quality control of raw sequencing data generated by multiple 3-prime-end sequencing protocols that include CEL-seq, MARS-seq, Chromium 10X and Drop-seq. scPipe produces a count matrix that is essential for downstream analysis along with an HTML report that summarises data quality. These results can be used as input for downstream analyses including normalization, visualization and statistical testing. scPipe performs this processing in a few simple R commands, promoting reproducible analysis of single-cell data that is compatible with the emerging suite of scRNA-seq analysis tools available in R/Bioconductor. The scPipe R package is available for download from https://www.bioconductor.org/packages/scPipe.

bioinformatics↗

bioRxiv2017

Deciphering Functional Redundancy in the Human Microbiome

Although the taxonomic composition of the human microbiome varies tremendously across individuals, its gene composition or functional capacity is highly conserved1-5---implying an ecological property known as functional redundancy. Such functional redundancy is thought to underlie the stability and resilience of the human microbiome6,7, but its origin is elusive. Here, we investigate the basis for functional redundancy in the human microbiome by analyzing its genomic content network --- a bipartite graph that links microbes to the genes in their genomes. We show that this network exhibits several topological features, such as highly nested structure and fat-tailed gene degree distribution, which favor high functional redundancy. To explain the origins of these topological features, we develop a simple genome evolution model that explicitly considers selection pressure, and the processes of gene gain and loss, and horizontal gene transfer. We find that moderate selection pressure and high horizontal gene transfer rate are necessary to generate genomic content networks with both highly nested structure and fat-tailed gene degree distribution, and consequently favor high functional redundancy. These findings provide insights into the relationships between structure and function in complex microbial communities. This work elucidates the potential ecological and evolutionary processes that create and maintain functional redundancy in the human microbiome and contribute to its resilience.

microbiology↗