Biology subjects

Wei, Y.

Publications and source records attributed to Wei, Y..

At least 289 records · Page 16Linked to original sources

bioRxiv2019

A novel machine learning based approach for iPS progenitor cell identification

Identification of induced pluripotent stem (iPS) progenitor cells, the iPS forming cells in early stage of reprogramming, could provide valuable information for studying the origin and underlying mechanism of iPS cells. However, it is very difficult to identify experimentally since there are no biomarkers known for early progenitor cells, and only about 6 days after reprogramming initiation, iPS cells can be experimentally determined via fluorescent probes. What is more, the ratio of progenitor cells during early reprograming period is below 5%, which is too low to capture experimentally in the early stage.\n\nIn this paper, we propose a novel computational approach for the identification of iPS progenitor cells based on machine learning and microscopic image analysis. Firstly, we record the reprogramming process using a live cell imaging system after 48 hours of infection with retroviruses expressing Oct4, Sox2 and Klf4, later iPS progenitor cells and normal murine embryonic fibroblasts (MEFs) within 3 to 5 days after infection are labeled by retrospectively tracing the time-lapse microscopic image. We then calculate 11 types of cell morphological and motion features such as area, speed, etc., and select best time windows for modeling and perform feature selection. Finally, a prediction model using XGBoost is built based on the selected six types of features and best time windows. Our model allows several missing values/frames in the sample datasets, thus it is applicable to a wide range of scenarios.\n\nCross-validation, holdout validation and independent test experiments showed that the minimum precision is above 52%, that is, the ratio of predicted progenitor cells within 3 to 5 days after viral infection is above 52%. The results also confirmed that the morphology and motion pattern of iPS progenitor cells is different from that of normal MEFs, which helps with the machine learning methods for iPS progenitor cell identification.\n\nAuthor SummaryIdentification of induced pluripotent stem (iPS) progenitor cells could provide valuable information for studying the origin and underlying mechanism of iPS cells. However, it is very difficult to identify experimentally since there are no biomarkers known for early progenitor cells, and only after about 6 days of induction, iPS cells can be experimentally determined via fluorescent probes. What is more, the percentage of the progenitor cells during the early induction period is below 5%, too low to capture experimentally in early stage. In this work, we proposed an approach for the identification of iPS progenitor cells, the iPS forming cells, based on machine learning and microscopic image analysis. The aim is to help biologists to enrich iPS progenitor cells during the early stage of induction, which allows experimentalists to select iPS progenitor cells with much higher probability, and furthermore to study the biomarkers which trigger the reprogramming process.

bioinformatics↗

bioRxiv2019

Considering founding and variable genomes is critical in studying polyploid evolution

A wide range of differences between the subgenomes, termed as subgenome asymmetry or SA, has been reported in various polyploids and different species seem to have different responses to polyploidization. We compared subgenome differences in gene ratio and relative diversity between artificial and natural genotypes of several allopolyploid species. Surprisingly, consistent differences in neither gene ratio nor relative diversity between the subgenomes were detected between these two types of polyploid genotypes although they differ in times exposed to evolutional selection. As expected, the estimated ratio of retained genes between a subgenome and its diploid donor was invariably higher for the artificial allopolyploid genotypes due likely to the presence of variable genome components (VGC). Clearly, the presence of VGC means that exaggerated differences between a donor and a subgenome in a polyploid are inevitable when random genotypes were used to represent species of either a polyploid or its donors. SA was also detected in genotypes before the completion of the polyploidization events as well as in those which were not formed via polyploidization. Considering that significant changes during and following polyploidization have been detected in previous studies, our results suggest that the influence of VGC needs to be considered in evaluating SA and that diploid donors may define changes in polyploid evolution.

evolutionary biology↗

bioRxiv2019

How do three cytosolic glutamine synthetase isozymes of wheat perform N assimilation and translocation?

To understand how the three cytosolic glutamine synthetase (GS1) isozymes of wheat (Triticum aestivum L., TaGS1) perform nitrogen assimilation and translocation, we studied the kinetic properties of TaGS1 isozymes, the effects of nitrogen on the expression and localization of TaGS1 isozymes with specific antibodies, and the nitrogen metabolism. The results showed TaGS1;1, the dominant TaGS1 isozyme, had a high affinity for substrates, and was widely localized in the mesophyll cells, root pericycle and root tip meristematic zone, suggesting it was the primary isozyme for N assimilation. TaGS1;2, with a high affinity for Glu, was activated by Gln, and was mainly localized in the around vascular tissues, indicating that TaGS1;2 catalyzed Gln synthesis in low Glu concentration, then the Gln returned to activate TaGS1;2, which may lead to the rapid accumulation of Gln around the vascular tissues. TaGS1;3 had low affinity for substrates but the highest Vmax among TaGS1, was mainly localized in the root tip meristematic zone; exogenous NH4+ could promote TaGS1;3 expressing, indicating that TaGS1;3 could rapidly assimilate NH4+ to relieve NH4+ toxicity. In conclusion, TaGS1;1, TaGS1;2 and TaGS1;3 have different role in N assimilation, Gln translocation and relieving ammonium toxicity, respectively, and synergistically perform nitrogen assimilation and translocation.\n\nHighlightThree cytosolic glutamine synthase isozymes of wheat have different role and synergistically perform nitrogen assimilation and translocation.

plant biology↗

bioRxiv2019

Polarized Lung Inflammation and Tie2/Angiopoietin-Mediated Endothelial Dysfunction during Severe Orientia tsutsugamushi Infection

Orientia tsutsugamushi infection can cause acute lung injury and high mortality in humans; however, the underlying mechanisms are unclear. Here, we tested a hypothesis that dysregulated pulmonary inflammation and Tie2-mediated endothelial malfunction contribute to lung damage. Using a murine model of lethal O. tsutsugamushi infection, we demonstrated pathological characteristics of vascular activation and tissue damage: 1) a significant increase of ICAM-1, VEGFR2, and angiopoietin-2 (Ang2) proteins in inflamed tissues and lung-derived endothelial cells (EC), 2) a progressive loss of endothelial quiescent and junction proteins (Ang1, VE-cadherin/CD144, occuludin), and 3) a profound impairment of Tie2 receptor at the transcriptional and functional levels. In vitro infection of primary human EC cultures and serum Ang2 proteins in scrub typhus patients support our animal studies, implying endothelial dysfunction in severe scrub typhus. Flow cytometric analyses of lung-recovered cells further revealed that pulmonary macrophages (M{Phi}) were polarized toward an M1-like phenotype (CD80+CD64+CD11b+Ly6G-) during the onset of disease and prior to host death, which correlated with the significant loss of CD31+CD45- ECs and M2-like (CD206+CD64+CD11b+Ly6G-) cells. In vitro studies indicated extensive bacterial replication in M2-type, but not M1-type, M{Phi}s, implying the protective and pathogenic roles of M1-skewed responses. This is the first detailed investigation of lung cellular immune responses during acute O. tsutsugamushi infection. It uncovers specific biomarkers for vascular dysfunction and M1-skewed inflammatory responses, highlighting future therapeutic research for the control of this neglected tropical disease.\n\nAuthor SummaryScrub typhus is a life-threatening disease, infecting an estimated one million people yearly. Acute lung injury is the most common clinical observation; however, its pathogenic biomarkers and mechanisms of progression remain unknown. Here, we used a lethal infection mouse model that parallels certain aspects of severe scrub typhus, primary human endothelial cell cultures, and patient sera to define pathogenic biomarkers following Orientia tsutsugamushi infections. We found a significant increase in the levels of endothelial activation/stress markers (angiopoietins and ICAM-1) in infected mouse lungs and in patient sera, but a progressive loss of endothelium-specific Tie2 receptor and junction proteins (VE-cadherin), at severe stages of disease. These signs of vasculature disruption positively correlated with the timing and magnitude of recruitment/activation of proinflammatory M{Phi} subsets in infected lungs. Bacterial growth in vitro was favored in M2-like, but not in M1-like, M{Phi}s. This study, for the first time, reveals endothelial malfunction and dysregulated inflammatory responses, suggesting potential therapeutic targets to ameliorate tissue damage and pathogenesis.

microbiology↗

bioRxiv2019

Analysis of the Mutant Selection Window and Killing of Mycoplasma hyopneumoniae for Doxycycline, Tylosin, Danofloxacin, Tiamulin, and Valnemulin

Mycoplasma hyopneumoniae is the major pathogenic microorganism causing enzootic pneumonia in pigs. With increasing resistance of M. hyopneumoniae to conventional antibiotics, treatment is becoming complicated. Herein, we investigated the mutant selection window (MSW) of doxycycline, tylosin, danofloxacin, tiamulin, and valnemulin for treating M. hyopneumoniae strain (ATCC 25934) to determine the likelihood of promoting resistance with continued use of these antibiotics. Minimum inhibitory concentration (MIC) values against M. hyopneumoniae were determined for each antimicrobial agent and ranged from 105 colony-forming units (CFU)/mL to 109 CFU/mL based on microdilution broth and agar dilution methods. The minimal concentration inhibiting colony formation by 99% (MIC99) and the mutant prevention concentration (MPC) were determined by the agar dilution method with three inoculum sizes. Antimicrobial killing was determined based on MIC99 and MPC values for all five agents. MIC values ranged from 0.001 to 0.25 g/mL based on the microdilution broth method, and from 0.008 to 1.0 g/mL based on the agar dilution method. MPC values ranged from 0.0016 to 10.24 g/mL. MPC/MIC99 values were ordered tylosin >doxycycline >danofloxacin >tiamulin >valnemulin. MPC achieved better bactericidal action than MIC99. Based on pharmacodynamic analyses, danofloxacin, tylosin, and doxycycline are more likely to select resistant mutants than tiamulin and valnemulin.

pharmacology and toxicology↗

bioRxiv2019

Genetic correlates of evolutionary adaptations in cognitive functional brain networks and their relationship to human cognitive functioning and disease

Cognitive functional networks such as the default-mode network (DMN), frontal-parietal network (FPN), and salience network (SN), are key networks of the human brain. Here, we show that the distinct rapid evolutionary cortical expansion of cognitive networks in the human brain, and most pronounced the DMN, runs parallel with high expression of genes important for human evolution (so-called HAR genes). Comparative gene expression examination then shows that HAR genes are more differentially expressed in cognitive networks in humans compared to the chimpanzee and macaque. Genes with distinct high expression in the DMN display broad involvement in the formation of synapses and dendrites. Next, we performed a genome-wide association analysis on functional MRI data, and show that HAR genes are associated with individual variations in DMN functional connectivity in todays human population. Finally, gene-set analysis suggests associations of HAR genes with intelligence, social cognition, and mental conditions such as schizophrenia and autism. Taken together, our results indicate that the expansion of higher-order functional networks and their cognitive properties have been an important locus of change in recent human brain evolution.

neuroscience↗

bioRxiv2019

Recombinant Expression of Aldehyde Dehydrogenase 2 (ALDH2) in Escherichia coli Nissle 1917 for Oral Delivery in ALDH2-Deficient Individuals

Turning red after consuming alcohol may seem like a mere social inconvenience. Yet, this flushing response is caused by an accumulation of acetaldehyde, a carcinogenic intermediate of alcohol metabolism. Aldehyde dehydrogenase 2 (ALDH2) deficiency, the result of a point mutation, produces a less efficient ALDH2. The resulting accumulation of acetaldehyde greatly increases the risk of developing esophageal and head and neck cancers. In this study, we produced recombinant ALDH2 in the probiotic E. coli Nissle 1917, which successfully reduces acetaldehyde levels in simulated oral conditions. Packaged in a hard candy, the ALDH2-probiotic would remain in the mouth to specifically target salivary acetaldehyde. Using mathematical modeling, we also determined how much recombinant ALDH2 is needed to reduce elevated acetaldehyde levels.\n\nFinancial DisclosureThis work was funded by Taipei American School. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.\n\nCompeting InterestsThe authors have declared that no competing interests exist.\n\nEthics StatementN/A\n\nData AvailabilityYes - all data are fully available without restriction. Sequences for the plasmids used in this study are available through the Registry of Standard Biological Parts. Links to raw data are included in Supplementary Information.

synthetic biology↗

bioRxiv2019

Stimulation augments spike sequence replay and memory consolidation during slow-wave sleep

Newly acquired memory traces are spontaneously reactivated during slow-wave sleep (SWS), leading to the consolidation of recent memories. Empirical studies found that sensory stimulation during SWS selectively enhances memory consolidation and the effect depends on the phase of stimulation. In this new study, we aimed to understand the mechanisms behind the role of sensory stimulation on memory consolidation using computational models implementing effects of neuromodulators to simulate transitions between awake and SWS sleep, and synaptic plasticity to allow the change of synaptic connections due to the training in awake or replay during sleep. We found that when closed-loop stimulation was applied during the Down states (900-2700) of sleep slow oscillation, particularly right before transition from Down to Up state, it significantly affected the spatio-temporal pattern of the slow-waves and maximized memory replay. In contrast, when the stimulation was presented during the Up states (2700-3600 and 00-900), it did not have a significant impact on the slow-waves or memory performance after sleep. For multiple memories trained in awake, presenting stimulation cues associated with specific memory trace could selectively augment replay and enhance consolidation of that memory and interfere with consolidation of the others (particularly weak) memories. Our study proposes a synaptic level mechanism of how memory consolidation is affected by sensory stimulation during sleep.\n\nSignificance statementStimulation, such as training-associated cues or auditory stimulation, during sleep can augment consolidation of the newly encoded memories. In this study, we used a computational model of the thalamocortical system to describe the mechanisms behind the role of stimulation in memory consolidation during slow-wave sleep. Our study suggested that stimulation preferentially strengthens the memory traces when delivered at specific phase of slow oscillations just before Down to Up state transition when it makes the largest impact on the spatio-temporal pattern of sleep slow waves. In the presence of multiple memories, presenting sensory cues during sleep could selectively strengthen selected memories. Our study proposes a synaptic level mechanism of how memory consolidation is affected by sensory stimulation during sleep.

neuroscience↗

bioRxiv2019

Shared and Distinct Dysfunction of Dynamic Connectivity Networks across Schizophrenia, Bipolar Disorder and Major Depression Disorder

Dynamic functional connectivity (DFC) analysis can capture time-varying properties of connectivity and may provide further information about transdiagnostic psychopathology across major psychiatric disorders. In this study, we used resting state functional MRI and a sliding-window method to study DFC in 150 schizophrenia (SZ), 100 bipolar disorder(BD), 150 major depressive disorder (MDD), and 210 healthy controls (HC). DFC were clustered into two functional connectivity states. Significant 4-group differences in DFC were found only in state 2. Post hoc analyses showed that transdiagnostic dysconnectivity among there disorders featured decreased connectivity within visual, somatomotor, salience and frontoparietal networks. Our results suggest that decreased connectivity within both lower-order (visual and somatomotor) and higher-order (salience and frontoparietal) networks may serve as transdiagnostic marker of these disorders, and that these dysconnectivity is state-dependent. Targeting these dysconnectivity may improve assessment and treatment for patients that having more than one of these disorders at the same time.

neuroscience↗

bioRxiv2019

Ginsenoside Rg1 defenses PC-12 cells against hydrogen peroxide-caused damage via up-regulation of miR-216a-5p

BackgroundSpinal cord injury (SCI) is a destructive trauma accompanying with local injury, half of which cause chronic paralysis. Ginsenoside Rg1 exerts anti-apoptosis and anti-autophagy properties. Therefore, our goal was to study the protective mechanism of Rg1 in attenuating cell injury.\n\nMethodsMiR-216a-5p inhibitor was transfected into PC-12 cells, then cells were pre-treated by Rg1 and treated with 300 M hydrogen peroxide (H2O2) for 24 h. CCK-8 and apoptosis experiments were done to test cell activity and apoptosis respectively. Expression of miR-216a-5p and cell damage relative factors was tested via qRT-PCR and western blot experiments, respectively.\n\nResultsH2O2 induced cell activity suppression, apoptosis and autophagy well at the concentration of 300 M, leading cell injury. Rg1 could attenuate cell injury induced by H2O2 at the working concentration of 200 M that it elevated cell activity, attenuated apoptosis and autophagy and activated PI3K/AKT and AMPK signal pathways. Further, miR-216a-5p was up-regulated by Rg1. Rg1 played its role in relieving cell injury by positively regulating miR-216a-5p.\n\nConclusionOur study demonstrated that Rg1 attenuated H2O2-caused cell injury through positively regulated miR-216a-5p.

biochemistry↗

bioRxiv2019

Homoploid hybridization signals due to ancestral subdivision: a case study on the D lineage in wheat

Homoploid hybrid speciation has been reported in a wide range of species since the exploitation of genome sequences in evolutionary studies. However, the interference of ancestral subdivision has not been adequately considered in many such investigations. Using the D lineage in wheat as an example, we showed clearly that ancestral subdivision has led to false detection of homoploid hybridization signals. We develop a novel statistical framework by examining the changes in shared ancestral variations and infer on the likelihood of speciation due to genuine homoploid hybridization or ancestral subdivisions. Applying this to wheat data, we found that homoploid hybridization was not involved in the origin of the D lineage contrary to the now widely held belief. This example indicates that the significance of homoploid hybrid speciation is likely exaggerated. The underlying methodology developed in this study should be valuable for clarifying whether homoploid hybridization has contributed to the speciation of many other species.

evolutionary biology↗

bioRxiv2019

GWAS of brain volume on 54,407 individuals and cross-trait analysis with intelligence identifies shared genomic loci and genes

The phenotypic correlation between human intelligence and brain volume (BV) is considerable (r{approx}0.40), and has been shown to be due to shared genetic factors1. To further examine specific genetic factors driving this correlation, we present genomic analyses of the genetic overlap between intelligence and BV using genome-wide association study (GWAS) results. First, we conducted the largest BV GWAS meta-analysis to date (N=54,407 individuals), followed by functional annotation and gene-mapping. We identified 35 genomic loci (27 novel), implicating 362 genes (346 novel) and 23 biological pathways for BV. Second, we used an existing GWAS for intelligence (N=269,867 individuals2), and estimated the genetic correlation (rg) between BV and intelligence to be 0.23. We show that the rg is driven by physical overlap of GWAS hits in 5 genomic loci. We identified 67 shared genes between BV and intelligence, which are mainly involved in important signaling pathways regulating cell growth. Out of these 67 we prioritized 32 that are most likely to have functional impact. These results provide new information on the genetics of BV and provide biological insight into BVs shared genetic etiology with intelligence.

genetics↗

bioRxiv2019

Targeting against HIV/HCV Co-infection using Machine Learning-based multitarget-quantitative structure-activity relationships (mt-QSAR) Methods

Co-infection between HIV-1 and HCV is common today in certain populations. However, treatment of co-infection is full of challenges with special consideration for potential hepatic safety and drug-drug interactions. Multitarget inhibitors with less toxicity may provide a promising therapeutic strategy for HIV/HCV co-infection. However, identification of one molecule acting on multiple targets simultaneously by experimental evaluation is costly and time-consuming. In silico target prediction tools provide more opportunities for the development of multitarget inhibitors. In this study, by combining naive Bayesian (NB) and support vector machine (SVM) algorithms with two types of molecular fingerprints (MACCS and ECFP6), 60 classification models were constructed to predict the active compounds toward 11 HIV-1 targets and 4 HCV targets based on the multitarget-quantitative structure-activity relationships (mt-QSAR). 5-fold cross-validation and test set validation was performed to confirm the performance of 60 classification models. Our results show that 60 mt-QSAR models appeared to have high classification accuracy in terms of ROC-AUC values ranging from 0.83 to 1 with a mean value of 0.97 for HIV-1 models, and ROC-AUC values ranging from 0.84 to 1 with a mean value of 0.96 for HCV. Furthermore, the 60 models were applied to comprehensively predict the potential targets for additional 46 compounds including 27 approved HIV-1 drugs, 10 approved HCV drugs and 9 selected compounds known to be active on one or more targets of HIV-1 or those of HCV. Finally, 18 hits including 7 HIV-1 approved drugs, 4 HCV approved drugs and 7 compounds were predicted to be HIV/HCV co-infection multitarget inhibitors. The reported bioactivity data confirmed that 7 compounds actually interacted with HIV-1 and HCV targets simultaneously with diverse binding affinities. Of those remaining predicted hits and chemical-protein interaction pairs involving the potential ability to suppress HIV/HCV co-infection deserve further investigation by experiments. This investigation shows that the mt-QSAR method is available to predict chemical-protein interaction for discovering multitarget inhibitors and provide a unique perspective on HIV/HCV co-infection treatment.

bioinformatics↗

bioRxiv2019

The hourglass organization of the C. elegans connectome

We approach the C. elegans connectome as an information processing network that receives input from about 90 sensory neurons, processes that information through a highly recurrent network of about 80 interneurons, and it produces a coordinated output from about 120 motor neurons that control the nematodes muscles. We focus on the feedforward flow of information from sensory neurons to motor neurons, and apply a recently developed network analysis framework referred to as the "hourglass effect". The analysis reveals that this feedforward flow traverses a small core ("hourglass waist") that consists of 10-15 interneurons. These are mostly the same interneurons that were previously shown (using a different analytical approach) to constitute the "rich-club" of the C. elegans connectome. This result is robust to the methodology that separates the feedforward from the feedback flow of information. The set of core interneurons remains mostly the same when we consider only chemical synapses or the combination of chemical synapses and gap junctions. The hourglass organization of the connectome suggests that C. elegans has some similarities with encoder-decoder artificial neural networks in which the input is first compressed and integrated in a low-dimensional latent space that encodes the given data in a more efficient manner, followed by a decoding network through which intermediate-level sub-functions are combined in different ways to compute the correlated outputs of the network. The core neurons at the hourglass waist represent the information bottleneck of the system, balancing the representation accuracy and compactness (complexity) of the given sensory information. Author SummaryThe C. elegans nematode is the only species for which the complete wiring diagram ("connectome") of its neural system has been mapped. The connectome provides architectural constraints that limit the scope of possible functions of a neural system. In this work, we identify one such architectural constraint: the C. elegans connectome includes a small set (10-15) of neurons that compress and integrate the information provided by the much larger set of sensory neurons. These intermediate-level neurons encode few sub-functions that are combined and re-used in different ways to activate the circuits of motor neurons, which drive all higher-level complex functions of the organism such as feeding or locomotion. We refer to this encoding-decoding structure as "hourglass architecture" and identify the core neurons at the "waist" of the hourglass. We also discuss the similarities between this property of the C. elegans connectome and artificial neural networks. The hourglass architecture opens a new way to think about, and experiment with, intermediate-level neurons between input and output neural circuits.

neuroscience↗

bioRxiv2019

The gut microbiota regulates mouse biliary regenerative responses

Injury to the biliary epithelium triggers cholangiopathies. However, factors involved in regulating the pathogenesis remains incompletely understood. Here, we report that the gut microbiota is important in regulating hepatobiliary fibroproliferative regenerative process. In helminth Clonorchis sinensis-induced bile duct injury model, wild-type mice showed more extensive peribiliary fibroproliferative responses than non-littermate IL-33-deficient mice. However, these reactions could be attenuated by co-housing of the animals together. In the meantime, the relatively fibroproliferative-resistant IL-33-deficient mice could become fibroproliferative-responsive especially in large intrahepatic bile duct by antibiotics treatment. Furthermore, microbiota-derived metabolite butyrate was able to inhibit biliary organoid expansion in vitro and temper 3,5-diethoxycarbonyl-1,4-dihydrocollidine-induced biliary fibrosis in vivo. Together, our data implies a potential way of management of hepatobiliary diseases by modulating gut microbiota.

microbiology↗

bioRxiv2019

A novel QTL conferring Fusarium crown rot resistance located on chromosome arm 6HL in barley

Fusarium crown rot (FCR), caused primarily by Fusarium pseudograminearum, is a devastating disease for cereal production in semi-arid regions worldwide. To identify and characterize loci conferring FCR resistance, we assessed a landrace AWCS799 which is among the top lines identified from a screening. Genetic control of its resistance was investigated by generating and analysing two populations of recombinant inbred lines. One of the populations was used for QTL detection and the other for validation. A novel QTL, located on the long arm of chromosome 6H (designated as Qcrs.caf-6H), was consistently detected in each of the four trials conducted against the mapping population. The QTL explained up to 29.1% of the phenotypic variance and its effect was confirmed in the validation population. Significant interactions between this resistance locus and either plant height or heading date were not detected, further facilitating its manipulation in breeding programs. Key MessageThis study identified and validated a novel and large-effect QTL conferring Fusarium crown rot resistance on the long arm of chromosome 6HL in barley.

plant biology↗

bioRxiv2019

Flexible Experimental Designs for Valid Single-cell RNA-sequencing Experiments Allowing Batch Effects Correction

Despite their widespread applications, single-cell RNA-sequencing (scRNA-seq) experiments are still plagued by batch effects and dropout events. Although the completely randomized experimental design has frequently been advocated to control for batch effects, it is rarely implemented in real applications due to time and budget constraints. Here, we mathematically prove that under two more flexible and realistic experimental designs--the "reference panel" and the "chain-type" designs--true biological variability can also be separated from batch effects. We develop Batch effects correction with Unknown Subtypes for scRNA-seq data (BUSseq), which is an interpretable Bayesian hierarchical model that closely follows the data-generating mechanism of scRNA-seq experiments. BUSseq can simultaneously correct batch effects, cluster cell types, impute missing data caused by dropout events, and detect differentially expressed genes without requiring a preliminary normalization step. We demonstrate that BUSseq outperforms existing methods with simulated and real data.

bioinformatics↗

bioRxiv2018

A Computational Solution To Improve Biomarker Reproducibility During Long-term Projects

Biomarkers are fundamental to basic and clinical research outcomes by reporting host responses and providing insight into disease pathophysiology. Measuring biomarkers with research-use ELISA kits is universal, yet lack of kit standardization and unexpected lot-to-lot variability presents analytic challenges for long-term projects. During an ongoing two-year project measuring plasma biomarkers in cancer patients, control concentrations for one biomarker (PF) decreased significantly after changes in ELISA kit lots. A comprehensive operations review pointed to standard curve shifts with the new kits, an analytic variable that jeopardized data already collected on hundreds of patient samples. After excluding other reasonable contributors to data variability, a computational solution was developed to provide a uniform platform for data analysis across multiple ELISA kit lots. The solution (ELISAtools) was developed within open-access R software in which variability between kits is treated as a batch effect. A defined best-fit Reference standard curve is modelled, a unique Shift factor \"S\" is calculated for every standard curve and data adjusted accordingly. The averaged S factors for PF ELISA kit lots #1-5 ranged from -0.086 to 0.735, and reduced control inter-assay variability from 62.4% to <9%, within quality control limits. S factors calculated for four other biomarkers provided a quantitative metric to monitor ELISAs over the 10 month study period for quality control purposes. Reproducible biomarker measurements are essential, particularly for long-term projects with valuable patient samples. Use of research-use ELISA kits is ubiquitous and judicious use of this computational solution maximizes biomarker reproducibility.

pathology↗