Biology subjects

Hu, M.

Publications and source records attributed to Hu, M..

At least 199 records · Page 11Linked to original sources

bioRxiv2020

Induction of hyperandrogenism and insulin resistance differentially modulates ferroptosis in uterine and placental tissues of pregnant rats

Ferroptosis, a form of regulated necrotic cell death, plays roles in diverse physiological processes and diseases. Women with polycystic ovary syndrome (PCOS) have hyperandrogenism and insulin resistance (HAIR) and an increased risk of miscarriage and placental dysfunction during pregnancy. However, whether maternal HAIR alters mechanisms leading to ferroptosis in the gravid uterus and placenta remains unknown. Previous studies in rats showed that maternal exposure to 5-dihydrotestosterone (DHT) and insulin (INS) from gestational day 7.5 to 13.5 induces HAIR and subsequently leads to placental insufficiency and fetal loss. We therefore hypothesized that maternal HAIR triggers ferroptosis in the uterus and placenta in association with fetal loss in pregnant rats. Compared with controls, we found that co-exposure to DHT and INS led to decreased levels of Gpx4 and glutathione (GSH), increased GSH+glutathione disulfide (GSSG) and malondialdehyde (MDA), aberrant expression of ferroptosis-associated genes (Acsl4, Tfrc, Slc7a11, and Gclc), increased iron deposition, and activated ERK/p38/JNK phosphorylation in the gravid uterus. However, in the placenta, DHT and INS exposure only partially altered the expression of ferroptosis-related markers (e.g., region-dependent Gpx4, GSH+GSSG, MDA, Gls2 and Slc7a11 mRNAs, and phosphorylated p38 levels). In the uteri co-exposed to DHT and INS, we also observed shrunken mitochondria with electron-dense cristae, which are key features of ferroptosis-related mitochondrial morphology, as well as increased expression of Dpp4, a mitochondria-encoded gene responsible for ferroptosis induction. In contrast, in placentas co-exposed to DHT and INS we found decreased expression of Dpp4 mRNA and increased expression of Cisd1 mRNA (a mitochondria-encoded iron-export factor). Further, DHT+INS-exposed pregnant rats exhibited decreased apoptosis in the uterus and increased necroptosis in the placenta. Our findings suggest that maternal HAIR causes the activation of ferroptosis in the gravid uterus and placenta, although this is mediated via different mechanisms operating at the molecular and cellular levels. Furthermore, our data suggest other cell death pathways may play a role in coordinating or compensating for HAIR-induced ferroptosis when the gravid uterus and placenta are dysfunctional. Suppl Figures: https://doi.org/10.6084/m9.figshare.11794059.v5

physiology↗

bioRxiv2020

CTCF Promotes Long-range Enhancer-promoter Interactions and Lineage-specific Gene Expression in Mammalian Cells

Topologically associating domains (TAD) and insulated neighborhoods (INs) have been proposed to constrain enhancer-promoter communications to enable cell-type specific transcription programs, but recent studies show that disruption of TADs and INs resulted in relatively mild changes in gene expression profiles. To better understand the role of chromatin architecture in dynamic enhancer-promoter contacts and lineage-specific gene expression, we have utilized the auxin-inducible degron system to acutely deplete CTCF, a key factor involved in TADs and IN formation, in mouse embryonic stem cells (mESCs) and examined chromatin architecture and gene regulation during neural differentiation. We find that while CTCF depletion leads to global weakening of TAD boundaries and loss of INs, only a minor fraction of enhancer-promoter contacts are lost, affecting a small subset of genes. The CTCF-dependent enhancer-promoter contacts tend to be long-range, spanning hundreds of kilobases, and are established directly by CTCF binding to promoters. Disruption of CTCF binding at the promoter reduces enhancer-promoter contacts and transcription, while artificial tethering of CTCF to the promoter restores the enhancer-promoter contacts and gene activation. Genome-wide analysis of CTCF binding and gene expression across multiple mouse tissues suggests that CTCF-dependent promoter-enhancer contacts may regulate expression of additional mouse genes, particularly those expressed in the brain. Our results uncover both CTCF-dependent and independent enhancer-promoter contacts, and highlight a distinct role for CTCF in promoting enhancer-promoter contacts and gene activation in addition to its insulator function.

genomics↗

bioRxiv2020

Persistent cell proliferation signals correlates with increased glycolysis in tumor hypoxia microenvironment across cancer types

BackgroundAltered metabolism is a hallmark of cancer and glycolysis is one of the important factors promoting tumor development. Given that the absence of multi-sample big data research about glycolysis, the molecular mechanisms involved in glycolysis or the relationships between glycolysis and tumor microenvironment are not fully studied. Thus, a more comprehensive approach in a pan-cancer landscape may be needed. MethodsHere, we develop a computational pipeline to study multi-omics molecular features defining glycolysis activity and identify molecular alterations that correlate with glycolysis. We apply a 22-gene expression signature to define the glycolysis activity landscape and verify the robustness using clinically defined glycolysis samples from several previous studies. Based on gene expression signature, we classify about 5552 of 9229 tumor samples into glycolysis score-high and score-low groups across 25 cancer types from The Cancer Genome Atlas (TCGA) and demonstrate their prognostic associations. Moreover, using genomes and transcriptome data, we characterize the association of copy-number aberrations (CNAs), somatic single-nucleotide variants (SNVs) and hypoxia signature with glycolysis activity. FindingsGene set variation analysis (GSVA) score by gene set expression was verified robustly to represent glycolytic activity and highly glycolytic tumors presented a poor overall survival in some cancer types. Then, we identified various types of molecular features promoting tumor cell proliferation were associated with glycolysis activity. Our study showed that TCA cycle and respiration electron transport were active in glycolysis-high tumors, indicating glycolysis was not a symptom of impaired oxidative metabolism. The glycolytic score significantly correlated with hypoxia score across all cancer types. Glycolysis score was also associated with elevated genomic instability. In all tumor types, high glycolysis tumors exhibited characteristic driver genes altered by CNAs identified multiple oncogenes and tumor suppressors. We observed widespread glycolysis-associated dysregulation of mRNA across cancers and screened out HSPA8 and P4HA1 as the potential modulating factor to glycolysis. Besides, the expression of genes encoding glycolytic enzymes positively correlated with genes in cell cycle. InterpretationThis is the first study to identify gene expression signatures that reflect glycolysis activity, which can be easily applied to large numbers of patient samples. Our analysis establishes a computational framework for characterizing glycolysis activity using gene expression data and defines correlation of glycolysis with the hypoxia microenvironment, tumor cell cycle and proliferation at a pan-cancer landscape. The findings suggest that the mechanisms whereby hypoxia influence glycolysis are likely multifactorial. Our finding is significant not just in demonstrating definition value for glycolysis but also in providing a comprehensive molecular-level understanding of glycolysis and suggesting a framework to guide combination therapy that may block the glycolysis pathway to control tumor growth in hypoxia microenvironment.

cancer biology↗

bioRxiv2020

Standardization and Harmonization of Distributed Multi-National Proteotype Analysis supporting Precision Medicine Studies

Cancer has no borders: Generation and analysis of molecular data across multiple centers worldwide is necessary to gain statistically significant clinical insights for the benefit of patients. Here we conceived and standardized a proteotype data generation and analysis workflow enabling distributed data generation and evaluated the quantitative data generated across laboratories of the international Cancer Moonshot consortium. Using harmonized mass spectrometry (MS) instrument platforms and standardized data acquisition procedures, we demonstrated robust, sensitive, and reproducible data generation across eleven sites in nine countries on seven consecutive days in a 24/7 operation mode. The data presented from the high-resolution MS1-based quantitative data-independent acquisition (HRMS1-DIA) workflow shows that coordinated proteotype data acquisition is feasible from clinical specimens using such standardized strategies. This work paves the way for the distributed multi-omic digitization of large clinical specimen cohorts across multiple sites as a prerequisite for turning molecular precision medicine into reality.

molecular biology↗

bioRxiv2020

3D Epigenomic Characterization Reveals Insights Into Gene Regulation and Lineage Specification During Corticogenesis

Lineage-specific epigenomic changes during human corticogenesis have previously remained elusive due to challenges with tissue heterogeneity and sample availability. Here, we analyze cis-regulatory chromatin interactions, open chromatin regions, and transcriptomes for radial glia, intermediate progenitor cells, excitatory neurons, and interneurons isolated from mid-gestational human brain samples. We show that chromatin looping underlies transcriptional regulation for lineage-specific genes, with transcription factor motifs, families of transposable elements, and disease-associated variants enriched at distal interacting regions in a cell type-specific manner. A subset of promoters exhibit unusually high degrees of chromatin interactivity, which we term super interactive promoters. Super interactive promoters are enriched for critical lineage-specific genes, suggesting that interactions at these loci contribute to the fine-tuning of cell type-specific transcription. Finally, we present CRISPRview, a novel approach for validating distal interacting regions in primary cells. Our study presents the first characterization of cell type-specific 3D epigenomic landscapes during human corticogenesis, advancing our understanding of gene regulation and lineage specification during human brain development.

genomics↗

bioRxiv2020

Long-term implicit memory for sequential auditory patterns in humans

To understand auditory scenes, listeners track and retain the statistics of sensory inputs as they unfold over time. We combined behavioural manipulation and modelling to investigate how sequence statistics are encoded into long-term memory and used to interpret incoming sensory signals. In a series of experiments, participants detected the emergence of regularly repeating patterns in novel rapid sound sequences. Unbeknownst to them, a few regular patterns reoccurred sparsely (every [~]3 minutes). Reoccurring sequences showed a rapidly growing detection time advantage over novel sequences. This effect was implicit, robust to interference, and persisted up to 7 weeks. Human performance was reproduced by a memory-constrained probabilistic model, where sequences are stored as n-grams and are subject to memory decay. Results suggest that similar psychological mechanisms may underlie integration processes over different-time scales in memory formation and flexible retrieval.

neuroscience↗

bioRxiv2020

DPHL: A pan-human protein mass spectrometry library for robust biomarker discovery using Data-Independent Acquisition and Parallel Reaction Monitoring

To answer the increasing need for detecting and validating protein biomarkers in clinical specimens, proteomic techniques are required that support the fast, reproducible and quantitative analysis of large clinical sample cohorts. Targeted mass spectrometry techniques, specifically SRM, PRM and the massively parallel SWATH/DIA technique have emerged as a powerful method for biomarker research. For optimal performance, they require prior knowledge about the fragment ion spectra of targeted peptides. In this report, we describe a mass spectrometric (MS) pipeline and spectral resource to support data-independent acquisition (DIA) and parallel reaction monitoring (PRM) based biomarker studies. To build the spectral resource we integrated common open-source MS computational tools to assemble an open source computational workflow based on Docker. It was then applied to generate a comprehensive DIA pan-human library (DPHL) from 1,096 data dependent acquisition (DDA) MS raw files, and it comprises 242,476 unique peptide sequences from 14,782 protein groups and 10,943 SwissProt-annotated proteins expressed in 16 types of cancer samples. In particular, tissue specimens from patients with prostate cancer, cervical cancer, colorectal cancer, hepatocellular carcinoma, gastric cancer, lung adenocarcinoma, squamous cell lung carcinoma, diseased thyroid, glioblastoma multiforme, sarcoma and diffuse large B-cell lymphoma (DLBCL), as well as plasma samples from a range of hematologic malignancies were collected from multiple clinics in China, the Netherlands and Singapore and included in the resource. This extensive spectral resource was then applied to a prostate cancer cohort of 17 patients, consisting of 8 patients with prostate cancer (PCa) and 9 with benign prostate hyperplasia (BPH), respectively. Data analysis of DIA data from these samples identified differential expressions of FASN, TPP1 and SPON2 in prostate tumors. Thereafter, PRM validation was applied to a larger PCa cohort of 57 patients and the differential expressions of FASN, TPP1 and SPON2 in prostate tumors were validated. As a second application, the DPHL spectral resource was applied to a patient cohort consisting of samples from 19 DLBCL patients and 18 healthy individuals. Differential expressions of CRP, CD44 and SAA1 between DLBCL cases and healthy controls were detected by DIA-MS and confirmed by PRM. These data demonstrate that the DPHL supported that DIA-PRM MS pipeline enables robust protein biomarker discoveries.

bioinformatics↗

bioRxiv2020

The intersection between the representation of the stimuli and the choice by neural ensembles in the primary visual cortex of the macaque.

Animals frequently make decisions based on sensory cues. In such a setting, the overlap in the information on the stimulus and on the choice is crucial for the formation of informed behavioral decisions. Yet, how the information on the stimulus and on the choice interact in the brain is poorly understood. Here, we study the representation of a binary decision variable in the primary visual cortex (V1) while macaque monkeys perform delayed match-to-sample task on naturalistic visual stimuli close to psychophysical threshold. Using population vectors, we demonstrate the overlap in decoding spaces on binary stimulus classes "match/non-match" and binary choices "same /different" of the animal. Leveraging this overlap, we use learning from the invariant information across the two classification problems to predict the choice of the animal as a time-dependent population signal. We show the importance of the across-neuron organization and the temporal structure of spike trains for the decision signal and suggest how noise correlations between neurons with similar decoding selectivity are helpful for the accumulation of the decision signal. Finally, we show that decision signal is primarily carried by bursting neurons in the superficial layers of the cortex. Author summaryV1 is necessary for normal visual processing and is known to process features of visual stimuli such as orientation, but whether V1 also encodes behavioral decisions is an unresolved issue, with conflicting evidence. Here, we demonstrate that V1 encodes a mixed variable that contains the information about the stimulus as well as about the choice. We learn the structure of population responses in trials pertaining to the variable "stimulus+choice", and apply the resulting population vectors to trials that differ only about the choice of the animal, but not about the stimulus class. Moreover, we learn structure of population responses on time-averaged data and then apply it on time-dependent (spiking) data. During the late phase of the trial, this procedure allows to predict the upcoming choice of the animal with a time-dependent population signal. The spiking signal of small neural population is sparse, and we hypothesize that positive correlations between neurons in the same decoding pool help the transmission of the decision-related information downstream. We find that noise correlations in the same decoding pool are significantly stronger than across coding pools, which corroborates our hypothesis on the benefit of noise correlations for the read-out of a time-dependent population signal.

neuroscience↗

bioRxiv2019

Multiplexed imaging of nucleome architectures in single cells of mammalian tissue

The three-dimensional architecture of the genome affects genomic functions. Multiple genome architectures at different length scales, including chromatin loops, domains, compartments, and regions associated with nuclear lamina and nucleoli, have been discovered. However, how these structures are arranged in the same cell and how they are correlated with each other in different cell types in mammalian tissue are largely unknown. Here, we developed Multiplexed Imaging of Nucleome Architectures that measures multiscale chromatin folding, copy numbers of numerous RNA species, and associations of numerous genomic regions with nuclear lamina, nucleoli and surface of chromosomes in the same, single cells. We applied this method in mouse fetal liver, and identified de novo cell-type-specific chromatin architectures associated with gene expression, as well as chromatin organization principles independent of cell type. Polymer simulation showed that both intra-chromosomal phase-separating interactions and extra-chromosomal interactions are necessary to establish the observed organization. Our experiments and modeling provide a multiscale and multi-faceted picture of chromatin folding and nucleome architectures in mammalian tissue and illustrate physical principles for maintaining chromatin organization.

cell biology↗

bioRxiv2019

Impact of temperature on Legionella pneumophila, its protozoan host cells, and the microbial diversity of the biofilm community of a pilot cooling tower

Legionella pneumophila (Lp) is a waterborne bacterium known for causing Legionnaires Disease, a severe pneumonia. Cooling towers are a major source of outbreaks, since they provide ideal conditions for Lp growth and produce aerosols. In such systems, Lp typically grow inside protozoan hosts. Several abiotic factors such as water temperature, pipe material and disinfection regime affect the colonization of cooling towers by Lp. The local physical and biological factors promoting the growth of Lp in water systems and its spatial distribution are not well understood. Therefore, we built a lab-scale cooling tower to study the dynamics of Lp colonization in relationship to the resident microbiota and spatial distribution. The pilot was filled with water from an operating cooling tower harboring low levels of Lp. It was seeded with Vermamoeba vermiformis, a natural host of Lp, and then inoculated with Lp. After 92 days of operation, the pilot was disassembled, the water was collected, and biofilm was extracted from the pipes. The microbiome was studied using 16S rRNA and 18S rRNA genes amplicon sequencing. The communities of the water and of the biofilm were highly dissimilar. The relative abundance of Legionella in water samples reached up to 11% whereas abundance in the biofilm was extremely low ([≤]0.5 %). In contrast, the host cells were mainly present in the biofilm. This suggest that Lp grows in host cells associated with biofilm and is then released back into the water following host cell lysis. In addition, water temperature shaped the bacterial and eukaryotic community of the biofilm, indicating that different parts of the systems may have different effects on Legionella growth.

microbiology↗

bioRxiv2019

Single cell lineage dynamics of the endosymbiotic cell type in a soft coral Xenia species

Many hard and soft corals harbor algae for photosynthesis. The algae live inside coral cells in a specialized membrane compartment called symbiosome, which shares the photosynthetically fixed carbon with coral host cells, while host cells provide inorganic carbon for photosynthesis1. This endosymbiotic relationship is critical for corals, but increased environmental stresses are causing corals to expel their endosymbiotic algae, i.e. coral bleaching, leading to coral death and degradation of marine ecosystem2. To date, the molecular pathways that orchestrate algal recognition, uptake, and maintenance in coral cells remain poorly understood. We report chromosome-level genome assembly of a fast-growing soft coral, Xenia species (sp.)3, and its use as a model to decipher the coral-algae endosymbiosis. Single cell RNA-sequencing (scRNA-seq) identified 13 cell types, including gastrodermis and cnidocytes, in Xenia sp. Importantly, we identified the endosymbiotic cell type that expresses a unique set of genes implicated in the recognition, phagocytosis/endocytosis, maintenance of algae, and host coral cell immune modulation. By applying scRNA-seq to investigate algal uptake in our new Xenia sp.. regeneration model, we uncovered a dynamic lineage progression from endosymbiotic progenitor state to mature endosymbiotic and post-endosymbiotic cell states. The evolutionarily conserved genes associated with the endosymbiotic process reported herein open the door to decipher common principles by which different corals uptake and expel their endosymbionts. Our study demonstrates the potential of single cell analyses to examine the similarities and differences of the endosymbiotic lifestyle among different coral species.

cell biology↗

bioRxiv2019

Label-free characterization of visual cortical areas in awake mice via three-photon microscopy reveals correlations between functional maps and structural substrates

Our understanding of the relationship between the structure and function of the intact brain is mainly shaped by magnetic resonance imaging. However, high resolution and deep-tissue imaging modalities are required to capture the subcellular relationship between structure and function, particularly in awake conditions. Here, we utilized a custom-made three-photon microscope to perform label-free third-harmonic generation (THG) microscopy as well as laser ablation to calculate effective attenuation lengths (EAL) of primary visual cortex and five adjacent visual cortical areas in awake mice. We identified each visual area precisely by retinotopic mapping via one-photon imaging of the calcium indicator GCaMP6s. EALs measured by depth-resolved THG microscopy in the cortex and white matter showed correspondence with the functional retinotopic sign map of each cortical area. To examine the basis for this correspondence, we used THG microscopy to examine several structural features of each visual area, including their cytoarchitecture, myeloarchitecture and blood vessel architecture. The cytoarchitecture of each area allowed us to estimate EAL values, which were comparable to experimental EAL values. The orientation of blood vessels and myelin fibers in the six areas were correlated with their EAL values. Ablation experiments, which provide ground truth measurements, generated 17 {+/-} 3 % longer EALs compared to those obtained with THG imaging but were consistent with the latter. These results demonstrate a strong correlation between structural substrates of visual cortical areas, represented by EALs, and their functional visual field representation maps.

neuroscience↗

bioRxiv2019

Robust Hi-C chromatin loop maps in human neurogenesis and brain tissues at high-resolution

Genome-wide mapping of chromatin interactions at high resolution remains experimentally and computationally challenging. Here we used a low-input \"easy Hi-C\" (eHi-C) protocol to map the 3D genome architecture in neurogenesis and brain tissues, and also developed an improved Hi-C bias-correction pipeline (HiCorr) enabling better identification of enhancer loops or aggregates at sub-TAD level. We compared ultra-deep 3D genome maps from 10 human tissue- or cell types, with a focus on stem cells and neural development. We found several large loci in skin-derived human iPSC lines showing recurrent 3D compartmental memory of somatic heterochromatin. Chromatin loop interactions, but not genome compartments, are hallmarks of neural differentiation. Interestingly, we observed many cell type- or differentiation-specific enhancer aggregates spanning large neighborhoods, supporting a phase-separation mechanism that stabilizes enhancer contacts during development. Finally, we demonstrated that chromatin loop outperforms eQTL in explaining neurological GWAS results, revealing a unique value of high-resolution 3D genome maps in elucidating the disease etiology.\n\nHighlightsO_LILow input \"easy Hi-C\" protocol compatible with 50-100K cells\nC_LIO_LIImproved Hi-C bias correction allows direct observation and accurate identification of sub-TAD chromatin loops and enhancer aggregates\nC_LIO_LIRecurrent architectural memory of somatic heterochromatin at compartment level in skin-derived hiPSCs\nC_LIO_LIChromatin loop, but not genome compartment, marks neural differentiation\nC_LIO_LIChromatin loop outperforms eQTL in defining brain GWAS target genes\nC_LI

genomics↗

bioRxiv2019

Unit quaternion description of spatial rotation in 3D electron cryo-microscopy

Electron cryo-microscopy (cryoEM) involves the estimation of orientations of projection images or three-dimensional (3D) volumes. However, the lack of statistical tools of rotations in cryoEM fails to answer the growing demands for adopting advanced statistical methods. In this study, we develop a comprehensive statistical tool specialized for cryoEM based on an unit quaternion description of spatial rotations. Some basic properties and definitions of the quaternion, as well as a way to use the unit quaternion to describe and perform rotations, are first recalled. Then, based on the unit quaternion, the distance and geodesic between rotations are designed for cryoEM to enable comparisons and interpolations between rotations, which are prerequisites of statistics of rotations in 3D cryoEM. Further, methods of directional statistics specialized for cryoEM are developed, including calculations of the average rotation, sampling, and inference with uniform and angular central Gaussian (ACG) distribution, as well as an estimation of the rotation precision. Finally, the method of handling molecular symmetry is introduced. Using the unit quaternion system for cryoEM, we provide comprehensive mathematical tools for the analysis of spatial rotations in cryoEM.

biophysics↗

bioRxiv2019

Long-lasting Analgesia via Targeted in vivo Epigenetic Repression of Nav1.7

Current treatments for chronic pain rely largely on opioids despite their unwanted side effects and risk of addiction. Genetic studies have identified in humans key targets pivotal to nociceptive processing, with the voltage-gated sodium channel, NaV1.7 (SCN9A), being perhaps the most promising candidate for analgesic drug development. Specifically, a hereditary loss-of-function mutation in NaV1.7 leads to insensitivity to pain without other neurodevelopmental alterations. However, the high sequence similarity between NaV subtypes has frustrated efforts to develop selective inhibitors. Here, we investigated targeted epigenetic repression of NaV1.7 via genome engineering approaches based on clustered regularly interspaced short palindromic repeats (CRISPR)-dCas9 and zinc finger proteins as a potential treatment for chronic pain. Towards this end, we first optimized the efficiency of NaV1.7 repression in vitro in Neuro2A cells, and then by the lumbar intrathecal route delivered both genome-engineering platforms via adeno-associated viruses (AAVs) to assess their effects in three mouse models of pain: carrageenan-induced inflammatory pain, paclitaxel-induced neuropathic pain and BzATP-induced pain. Our results demonstrate: one, effective repression of NaV1.7 in lumbar dorsal root ganglia; two, reduced thermal hyperalgesia in the inflammatory state; three, decreased tactile allodynia in the neuropathic state; and four, no changes in normal motor function. We anticipate this genomically scarless and non-addictive pain amelioration approach enabling Long-lasting Analgesia via Targeted in vivo Epigenetic Repression of Nav1.7, a methodology we dub pain LATER, will have significant therapeutic potential, such as for preemptive administration in anticipation of a pain stimulus (pre-operatively), or during an established chronic pain state.\n\nOne sentence summaryIn situ epigenome engineering approach for genomically scarless, durable, and non-addictive management of pain.

genetics↗

bioRxiv2019

TWO-SIGMA: a novel TWO-component SInGle cell Model-based Association method for single-cell RNA-seq data

In this paper, we develop TWO-SIGMA, a TWO-component SInGle cell Model-based Association method for differential expression (DE) analyses in single-cell RNA-seq (scRNA-seq) data. The first component models the probability of "drop-out" with a mixed-effects logistic regression model and the second component models the (conditional) mean expression with a mixed-effects negative binomial regression model. TWO-SIGMA is extremely flexible in that it: (i) does not require a log-transformation of the outcome, (ii) allows for overdispersed and zero-inflated counts, (iii) accommodates a correlation structure between cells from the same biological sample via random effect terms, (iv) can analyze unbalanced designs (in which the number of cells does not need to be identical for all samples), (v) can control for additional sample-level and cell-level covariates including batch effects, (vi) provides interpretable effect size estimates, and (vii) enables general tests of DE beyond two-group comparisons. To our knowledge, TWO-SIGMA is the only method for analyzing scRNA-seq data that can simultaneously accomplish each of these features. Simulations studies show that TWO-SIGMA outperforms alternative regression-based approaches in both type-I error control and power enhancement when the data contains even moderate within-sample correlation. A real data analysis using pancreas islet single-cells exhibits the flexibility of TWO-SIGMA and demonstrates that incorrectly failing to include random effect terms can have dramatic impacts on scientific conclusions. TWO-SIGMA is implemented in the R package twosigma available at https://github.com/edvanburen/twosigma.

genomics↗

bioRxiv2019

The structure of the population code in neural ensembles of V4 shapes pair-wise interactions on different time scales.

In visual areas of primates, neurons activate in parallel while the animal is engaged in a behavioral task. In this study, we examine the structure of the population code while the animal performs delayed match to sample task on complex natural images. The macaque monkeys visualized two consecutive stimuli that were either the same or different, while recorded with laminar arrays across the cortical depth in cortical areas V1 and V4. We decoded correct choice behavior from neural populations of simultaneously recorded units. Utilizing decoding weights, we divide neurons in most informative and less informative, and show that most informative neurons in V4, but not in V1, are more strongly synchronized, coupled and correlated than less informative neurons. As neurons are divided in two coding pools according to their coding preference, in V4, but not in V1, spiking synchrony, coupling and correlations within the coding pool are stronger than across coding pools. HighlightsO_LIIn a match-to-sample visual task, responses of neural populations in V1 and in V4 predict the stimulus class better than chance. C_LIO_LIIn V4, informative neurons are more strongly coupled, correlated and synchronized than less informative neurons. C_LIO_LIIn V4, neurons are more strongly coupled, correlated and synchronized within coding pools compared to across coding pools. C_LIO_LICorrelations within coding pools harm the performance of the classifier in both V1 and V4. C_LI

neuroscience↗

bioRxiv2019

Reading-out task variables as a low-dimensional reconstruction of neural spike trains in single trials.

We propose a new model of the read-out of spike trains that exploits the multivariate structure of responses of neural ensembles. Assuming the point of view of a read-out neuron that receives synaptic inputs from a population of projecting neurons, synaptic inputs are weighted with a heterogeneous set of weights. We propose that synaptic weights reflect the role of each neuron within the population for the computational task that the network has to solve. In our case, the computational task is discrimination of binary classes of stimuli, and weights are such as to maximize the discrimination capacity of the network. We compute synaptic weights as the feature weights of an optimal linear classifier. Once weights have been learned, they weight spike trains and allow to compute the post-synaptic current that modulates the spiking probability of the read-out unit in real time. We apply the model on parallel spike trains from V1 and V4 areas in the behaving monkey macaca mulatta, while the animal is engaged in a visual discrimination task with binary classes of stimuli. The read-out of spike trains with our model allows to discriminate the two classes of stimuli, while population PSTH entirely fails to do so. Splitting neurons in two subpopulations according to the sign of the weight, we show that population signals of the two functional subnetworks are negatively correlated. Disentangling the superficial, the middle and the deep layer of the cortex, we show that in both V1 and V4, superficial layers are the most important in discriminating binary classes of stimuli.

neuroscience↗