Biology subjects

Xu, J.

Publications and source records attributed to Xu, J..

At least 829 records · Page 46Linked to original sources

bioRxiv2019

Comparative analysis of diverse cell states establishes an epigenetic basis for inferring regulatory genes governing cell identity

Determining genes orchestrating cell differentiation in development and disease remains a fundamental goal of cell biology. This study establishes a genome-wide metric based on the gene-repressive tri-methylation of histone 3 lysine 27 (H3K27me3) across hundreds of diverse cell types to identify genetic regulators of cell differentiation. We introduce a computational method, TRIAGE, that uses discordance between gene-repressive tendency and expression to identify genetic drivers of cell identity. We apply TRIAGE to millions of genome-wide single-cell transcriptomes, diverse omics platforms, and eukaryotic cells and tissue types. Using a wide range of data, we validate TRIAGEs performance for identifying cell-type specific regulatory factors across diverse species including human, mouse, boar, bird, fish, and tunicate. Using CRISPR gene editing, we use TRIAGE to experimentally validate RNF220 as a regulator of Ciona cardiopharyngeal development and SIX3 as required for differentiation of endoderm in human pluripotent stem cells. A record of this papers Transparent Peer Review process is included in the Supplemental Information.

cell biology↗

bioRxiv2019

Efficient and accurate prediction of transmembrane topology from amino acid sequence only

MotivationRapid and accurate identification of transmembrane (TM) topology is well suited for the annotation of the entire membrane proteome. It is the initial step of predicting the structure and function of membrane proteins. However, existing methods that utilize only amino acid sequence information suffer from low prediction accuracy, whereas methods that exploit sequence profile or consensus need too much computational time.\n\nMethodHere we propose a deep learning framework DeepCNF that predicts TM topology from amino acid sequence only. Compared to previous sequence-based approaches that use hidden Markov models or dynamic Bayesian networks, DeepCNF is able to incorporate much more contextual information by a hierarchical deep neural network, while simultaneously modeling the interdependency between adjacent topology labels.\n\nResultExperimental results show that PureseqTM not only outperforms existing sequence-based methods, but also reaches or even surpasses the profile/consensus methods. On the 39 newly released membrane proteins, our approach successfully identifies the correct TM segments and boundaries for at least 3 cases while all existing methods fail to do so. When applied to the entire human proteome, our method can identify the incorrect annotations of TM regions by UniProt and discover the membrane-related proteins that are not manually curated as membrane proteins.\n\nAvailabilityhttp://pureseqtm.predmp.com/

bioinformatics↗

bioRxiv2019

Analysis of distance-based protein structure prediction by deep learning in CASP13

This paper reports the CASP13 results of distance-based contact prediction, threading and folding methods implemented in three RaptorX servers, which are built upon the powerful deep convolutional residual neural network (ResNet) method initiated by us for contact prediction in CASP12. On the 32 CASP13 FM (free-modeling) targets with a median MSA (multiple sequence alignment) depth of 36, RaptorX yielded the best contact prediction among 46 groups and almost the best 3D structure modeling among all server groups without time-consuming conformation sampling. In particular, RaptorX achieved top L/5, L/2 and L long-range contact precision of 70%, 58% and 45%, respectively, and predicted correct folds (TMscore>0.5) for 18 of 32 targets. Although on average underperforming AlphaFold in 3D modeling, RaptorX predicted correct folds for all FM targets with >300 residues (T0950-D1, T0969-D1 and T1000-D2) and generated the best 3D models for T0950-D1 and T0969-D1 among all groups. This CASP13 test confirms our previous findings: (1) predicted distance is more useful than contacts for both template-based and free modeling; and (2) structure modeling may be improved by integrating alignment and co-evolutionary information via deep learning. This paper will discuss progress we have made since CASP12, the strength and weakness of our methods, and why deep learning performed much better in CASP13.

bioinformatics↗

bioRxiv2019

Palmitoylation is critically required for cancer intrinsic PD-1 expression and functions

Programmed cell death protein 1 (PD-1) is a crucial anticancer target, but the relatively low response rate and acquired resistance to existing antibody drugs highlight an urgent need to develop alternative targeting strategies. Here we report the palmitoylation of PD-1, discovered the main DHHC enzyme for this modification, revealed the mechanism for its effect on PD-1 expression, and rationally developed a peptide for targeting PD-1 expression. Palmitoylation promoted the trafficking of PD-1 to recycling endosome, thus preventing its lysosome-dependent degradation. Palmitoylation was required for the activation of PD-1 downstream signaling, and targeting palmitoylation by pharmacological inhibitor or depleting the modification enzyme caused significant anti-tumor effects. A peptide was designed to competitively inhibit PD-1 palmitoylation and expression, opening a new route for developing PD-1 inhibitors as a strategy for cancer immunotherapy.\n\nSignificanceWe show for the first time that PD-1 is palmitoylated, identify DHHC9 as the predominant enzyme for its palmitoylation, and reveal the molecular mechanisms underlying its effects on PD-1 stability and functions. Importantly, we also designed a competitive inhibitor targeting PD-1 palmitoylation, and this first-in-class molecule may inspire the development of new checkpoint inhibitors.

cancer biology↗

bioRxiv2019

Sing cell RNA-seq reveals cellular diversity and developmental characteristics of human infant retina

Retina, located in the innermost layer of the eye of human, holds the decisive role in visual perception. Dissecting the heterogeneity of retina is essential for understanding the mechanism of vision formation and even the development of central nervous system (CNS). Here, we performed single cell RNA-seq, analyzed 57,832 cells from human infant donors, resulting in 20 distinct clusters representing major cell types in retina: rod photoreceptors, cone photoreceptors, bipolar cells, horizontal cells, amacrine cells, Muller glia cells and microglia. We next constructed extensive networks of intercellular communication and identified ligand-receptor interactions playing crucial roles in regulating neural cell development and immune homeostasis in retina. Though re-clustering, we identified known subtypes in cone PRs and additional unreported subpopulations and corresponding markers in rod PRs as well as bipolar cells. Additionally, we linked inherited retinal disease to certain cell subtypes or subpopulations through enrichment analysis. Intriguingly, we found that status and functions of photoreceptors changed drastically between early and late retina. Overall, our study offers the first retinal cell atlas in human infants, dissecting the heterogeneity of retina and identifying the key molecules in the developmental process, which provides an important resource that will pave the way for retina development mechanism research and regenerative medicine concerning retinal biology.

cell biology↗

bioRxiv2019

LFMD: a new likelihood-based method to detect low-frequency mutations without molecular tags

As next-generation sequencing (NGS) and liquid biopsy become more prevalent in research and in the clinic, there is an increasing need for better methods to reduce cost and improve sensitivity and specificity of low-frequency mutation detection (where the Alternative Allele Frequency, or AAF, is less than 1%). Here we propose a likelihood-based approach, called Low-Frequency Mutation Detector (LFMD), which combines the advantages of duplex sequencing (DS) and the bottleneck sequencing system (BotSeqS) to maximize the utilization of duplicate reads. Compared with the existing state-of-the-art methods, DS, Du Novo, UMI-tools, and Unified Consensus Maker, our method achieves higher sensitivity, higher specificity (< 4 x 10-10 errors per base sequenced) and lower cost (reduced by ~70% at best) without involving additional experimental steps, customized adapters or molecular tags. LFMD is useful in areas where high precision is required, such as drug resistance prediction and cancer screening. As an example of LFMDs applications, mitochondrial heterogeneity analysis of 28 human brain samples across different stages of Alzheimers Disease (AD) showed that the canonical oxidative damage related mutations, C:G>A:T, are significantly increased in the mid-stage group. This is consistent with the Mitochondrial Free Radical Theory of Aging, suggesting that AD may be linked to the aging of brain cells induced by oxidative damage.

bioinformatics↗

bioRxiv2019

Hypotonic stress induces fast, reversible degradation of the vimentin cytoskeleton via intracellular calcium release

The dynamic response of the cell to osmotic changes is critical to its physiology and has been widely exploited for cell manipulation. Using 3D-STORM super-resolution microscopy, here we examine the hypotonic stress-induced ultrastructural changes of the cytoskeleton of a common fibroblast cell type. Unexpectedly, we observe a fast, yet reversible dissolution of the vimentin intermediate filament system that precedes ultrastructural changes of the supposedly more dynamic actin and tubulin cytoskeletal systems, as well as changes in cell morphology. In combination with calcium imaging and biochemical analysis, we next show that the vimentin-specific fast cytoskeletal degradation under hypotonic stress is due to proteolysis by the calcium-dependent protease calpain. We find the process to be activated by the hypotonic stress-induced calcium release from intracellular stores, and so is efficiently suppressed by inhibiting any part of the IP3-Ca2+-calpain pathway we establish. Together, our findings highlight an unexpected, fast degradation mechanism for the vimentin cytoskeleton in response to the external stimuli, and point to the significant, yet previously overlooked physiological impacts of hypotonic stress-induced intracellular calcium release on cell ultrastructure and function.

cell biology↗

bioRxiv2019

VRmol: an Integrative Cloud-Based Virtual Reality System to Explore Macromolecular Structure

Past decades have seen much effort devoted to the tool development for biomacromolecule structural visualization and analysis. In recent years, we have witnessed tremendous advancement in computational technologies, especially in graphics and internet. Here we present an integrative cloud-based, virtual reality (VR) system -- VRmol that exploits these technology advances to visualize and study macromolecule structures in various forms. The immersive environment can improve user experiences in visualization, calculation, and editing of complex structures with more intuitive perception. VRmol is also a system with cloud-based capability to directly access disease-related databases, and couples the structure with genomic variation as well corresponding drugs information in a visual interface, thus serving as an integrative platform to aid structure-based translational research and drug design. VRmol is freely available at https://VRmol.net.

bioinformatics↗

bioRxiv2019

Predicting Drug Protein Interaction using Quasi-Visual Question Answering System

Identifying novel drug-protein interactions is crucial for drug discovery. For this purpose, many machine learning-based methods have been developed based on drug descriptors and one-dimensional (1D) protein sequences. However, protein sequence cant accurately reflect the interactions in 3D space. On the other hand, a direct input of 3D structure is of low efficiency due to the sparse 3D matrix, and is also prevented by limited number of co-crystal structures available for training. In this work, we propose an end-to-end deep learning framework to predict the interactions by representing proteins with 2D distance map from monomer structures (Image), and drugs with molecular linear notation (String), following the Visual Question Answering mode. For an efficient training of the system, we introduced a dynamic attentive convolutional neural network to learn fixed-size representations from the variable-length distance maps and a self-attentional sequential model to automatically extract semantic features from the linear notations. Extensive experiments demonstrate that our model obtains competitive performance against state-of-the-art baselines on the DUD-E, Human and BindingDB benchmark datasets. Further attention visualization provides biological interpretation to depict highlighted regions of both protein and drug molecules.

bioinformatics↗

bioRxiv2019

3.1A structure of yeast RNA polymerase II elongation complex stalled at a cyclobutane pyrimidine dimer lesion solved using streptavidin affinity grids

Despite significant advances in all aspects of single particle cryo-electron microscopy (cryo-EM), specimen preparation still remains a challenge. During sample preparation, macromolecules interact with the air-water interface, which often leads to detrimental effects such as denaturation or adoption of preferred orientations, ultimately hindering structure determination. Randomly biotinylating the protein of interest and then tethering it to a cryo-EM grid coated with two-dimensional crystals of streptavidin (acting as an affinity surface) can prevent the protein from interacting with the air-water interface. Recently, this approach was successfully used to solve a high-resolution structure of a test sample, a bacterial ribosome. However, the general applicability of this method to samples where interaction with the air-water interface is problematic remains to be determined. Here we report a 3.1[A] structure of a RNA polymerase II elongation complex stalled at a cyclobutane pyrimidine dimer lesion (Pol II EC(CPD)) solved using streptavidin grids. Our previous attempt to solve this structure using conventional sample preparation methods resulted in a poor quality cryo-EM map due to Pol II EC(CPD)s adopting a strong preferred orientation. Imaging the same sample on streptavidin grids led to a high-resolution structure with little anisotropy, showing that streptavidin affinity grids could be used as a general strategy to address challenges posed by interaction with the air-water interface.

molecular biology↗

bioRxiv2019

An ATAC-seq atlas of chromatin accessibility in mouse tissues

The Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) is a fundamental epigenomics approach and has been widely used in profiling the chromatin accessibility dynamics in multiple species. A comprehensive reference of ATAC-seq datasets for mammalian tissues is important for the understanding of regulatory specificity and developmental abnormality caused by genetic or environmental alterations. Here, we report an adult mouse ATAC-seq atlas by producing a total of 66 ATAC-seq profiles from 20 primary tissues of both male and female mice. The ATAC-seq read enrichment, fragment size distribution, and reproducibility between replicates demonstrated the high quality of the full dataset. We identified a total of 296,574 accessible elements, of which 26,916 showed tissue-specific accessibility. Further, we identified key transcription factors specific to distinct tissues and found that the enrichment of each motif reflects the developmental similarities across tissues. In summary, our study provides an important resource on the mouse epigenome and will be of great importance to various scientific disciplines such as development, cell reprogramming, and genetic disease.

genomics↗

bioRxiv2019

Plasma Circular RNA Panel to Diagnose Hepatitis B Virus-Related Hepatocellular Carcinoma

To explore whether plasma circular RNAs (circRNAs) can diagnose hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC), microarray and qPCR were used to identify plasma circRNAs that were increased in HCC patients compared with controls (including healthy controls, chronic hepatitis B, HBV-related liver cirrhosis and HCC patients). A logistic regression model was constructed using a training set (n=313) and then validated using another two independent sets (n=306 and 526, respectively). Area under the receiver operating characteristic curve (AUC) was used to evaluate diagnostic accuracy. We identified a plasma circRNA panel (CircPanel) containing three circRNAs (hsa_circ_0000976, hsa_circ_0007750 and hsa_circ_0139897) that could detect HCC. CircPanel showed a higher accuracy than AFP (alpha-fetoprotein) to distinguish individuals with HCC from controls in all three sets (AUC 0.863 [95% CI 0.819-0.907] vs 0.790 [0.738-0.842], P=0.036 in training set; 0.843 [0.796-0.890] vs 0.747 [0.691-0.804], P=0.011 in validation set 1 and 0.864 [0.830-0.898] vs 0.769 [0.728-0.810], P<0.001 in validation set 2). CircPanel also performed well in detecting Small-HCC (solitary, [≤]3cm), AFP-negative HCC and AFP-negative Small-HCC.\n\nSignificance of this studyWhat is already known about this subject? O_LIThe diagnostic accuracy of alpha-fetoprotein (AFP) in detecting hepatocellular carcinoma (HCC) is unsatisfactory.\nC_LIO_LICircular RNA (circRNA) expression profiles in HCC and adjacent nontumor liver tissues are significantly different.\nC_LIO_LIPlasma circRNAs are enriched, stable and can be biomarkers for various diseases.\nC_LI\n\nWhat are the new findings? O_LIThe expression of circRNAs in the plasma from HCC patients and chronic hepatitis B is significantly different.\nC_LIO_LIPlasma circRNA panel (CircPanel, including hsa_circ_0000976, hsa_circ_0007750 and hsa_circ_0139897) has a higher accuracy than AFP to distinguish individuals with HCC or Small-HCC (solitary, [≤]3cm) from controls (healthy controls, chronic hepatitis B and HBV-related liver cirrhosis).\nC_LIO_LICircPanel also performs well in diagnosing AFP-negative HCC and AFP-negative Small-HCC.\nC_LI\n\nHow might it impact on clinical practice in the foreseeable future? Plasma CircPanel can be a diagnostic biomarker in detecting HCC and improves the diagnostic accuracy.

cancer biology↗

bioRxiv2019

Genotype-free demultiplexing of pooled single-cell RNA-seq

A variety of experimental and computational methods have been developed to demultiplex samples from pooled individuals in a single-cell RNA sequencing (scRNA-Seq) experiment which either require adding information (such as hashtag barcodes) or measuring information (such as genotypes) prior to pooling. We introduce scSplit which utilises genetic differences inferred from scRNA-Seq data alone to demultiplex pooled samples. scSplit also extracts a minimal set of high confidence presence/absence genotypes in each cluster which can be used to map clusters to original samples. Using a range of simulated, merged individual-sample as well as pooled multi-individual scRNA-Seq datasets, we show that scSplit is highly accurate and concordant with demuxlet predictions. Furthermore, scSplit predictions are highly consistent with the known truth in cell-hashing dataset. We also show that multiplexed-scRNA-Seq can be used to reduce batch effects caused by technical biases. scSplit is ideally suited to samples for which external genome-wide genotype data cannot be obtained (for example non-model organisms), or for which it is impossible to obtain unmixed samples directly, such as mixtures of genetically distinct tumour cells, or mixed infections. scSplit is available at: https://github.com/jon-xu/scSplit

bioinformatics↗

bioRxiv2019

LonP1 Orchestrates UPRmt and UPRER and Mitochondrial Dynamics to Regulate Heart Function

Protein quality control is pivotal to cellular homeostasis and integrity of cardiomyocytes for maintenance of normal heart function. The unfolded protein response (UPR) is an adaptive process to modulate protein quality control in the endoplasmic reticulum (ER) and mitochondria, and is accordingly termed UPRER and UPRmt, respectively. Lon protease (LonP1) is a highly conserved mitochondrial protease to modulate UPRmt, which is involved in regulating metabolism, mitophagy, and stress response. However, whether LonP1 regulates UPRER remains elusive. To investigate the regulation of protein quality control in cardiomyocytes, we generated cardiac-specific LonP1 deletion mice. Our findings show that LonP1 deficiency caused impaired mitochondrial respiratory function and fragmentation. Surprisingly, both UPRER and UPRmt is substantially induced in LonP1-deletion heart suggesting of LonP1 as a novel regulator of UPRER; however, the activation of UPRER occurs earlier than UPRmt in response to LonP1 deletion. Consequently, cardiac-specific LonP1 deficiency causes aberrant metabolic reprogramming of cardiomyocytes, pathological heart remodeling, as well as impeded heart function. Thus, we uncovered the novel function of LonP1 as an UPRmt mediator, and reciprocal orchestration of UPRmt and UPRER and mitochondrial dynamics regulated by LonP1 in the cardiomyocytes that is critical to maintain heart function, which offers exciting new insights into the potential therapeutic strategy for heart failure.

cell biology↗

bioRxiv2019

An Integrated Framework for Genome Analysis Reveals Numerous Previously Unrecognizable Structural Variants in Leukemia Patients' Samples

While genomic analysis of tumors has stimulated major advances in cancer diagnosis, prognosis and treatment, current methods fail to identify a large fraction of somatic structural variants in tumors. We have applied a combination of whole genome sequencing and optical genome mapping to a number of adult and pediatric leukemia samples, which revealed in each of these samples a large number of structural variants not recognizable by current tools of genomic analyses. We developed computational methods to determine which of those variants likely arose as somatic mutations. The method identified 97% of the structural variants previously reported by karyotype analysis of these samples and revealed an additional fivefold more such somatic rearrangements. The method identified on average tens of previously unrecognizable inversions and duplications and hundreds of previously unrecognizable insertions and deletions. These structural variants recurrently affected a number of leukemia associated genes as well as cancer driver genes not previously associated with leukemia and genes not previously associated with cancer. A number of variants only affected intergenic regions but caused cis-acting alterations in expression of neighboring genes. Analysis of TCGA data indicates that the status of several of the recurrently mutated genes identified in this study significantly affect survival of AML patients. Our results suggest that current genomic analysis methods fail to identify a majority of structural variants in leukemia samples and this lacunae may hamper diagnostic and prognostic efforts.

genomics↗

bioRxiv2019

A Signaling Network based Computational Model to Uncover Loop as the Novel Molecular Mechanisms for Medulloblastoma

Medulloblastoma (MB) is the most common malignant brain tumor in children. Despite aggressive therapy, about one-third of patients with MB still die, and survivors suffer severe long-term side effects due to the treatments. The poor post-treatment outcomes are tightly linked to unpredictable drug resistance. Therefore, before developing robust single drug or drug combination recommendation algorithms, uncovering the underlying protein-protein interaction (PPI) network patterns that accurately explain and predict drug resistances for MB subtypes is essential and important. In this study, we hypothesize that the loop sub-structure within the PPI network can explain and predict drug resistance. Both static and dynamic models are built to evaluate this hypothesis for three MB subtypes. Specifically, a static model is created to first validate that many reported therapeutic targets are located topologically on highly deregulated loop sub-structure and then to characterize the loop for tumors without treatment. Next, with the after-treatment time-series genomics data, a dynamic hidden Markov model (HMM) with newly designed initialization scheme estimates the successful and unsuccessful occurrence probabilities for each given PPI and then re-delineates the loop for post-treatment tumors. Finally, the comparison of loop structures pre- and post-treatment distinguishes effective and ineffective treatment options, demonstrating that the loop sub-structure is capable of interpreting the mechanism of drug resistance. In summary, effective treatments show much stronger inhibition of cell cycle and DNA replication proteins when compared to ineffective treatments after considering the cross talk of multiple pathways (the loop).

bioinformatics↗

bioRxiv2019

Evaluations of short telomere risk-associated single nucleotide polymorphisms in telomerase reverse transcriptase gene on telomere length maintenance

Telomere biology disorders (TBDs) refer to a spectrum of tissue degenerative disorders caused by genetic mutations in telomere biology genes. Most patients with TBDs suffer from telomere maintenance defects secondary to telomerase deficiency. While the highly penetrant mutations in the telomerase reverse transcriptase (TERT) gene that drive disease onset and progression of TBDs are relatively rare, there exist several single nucleotide polymorphisms (SNPs) in TERT that have been linked to various diseases in the TBD spectrum. In this study, we investigated the biochemical properties of five TERT variants. In an ex vivo cell model, we found that primary human fibroblasts expressing nonsynonymous TERT SNPs had comparable cell growth kinetics to primary cells expressing WT-TERT, while a parallel vector control expressing-cell line entered replicative senescence. At the molecular level, primary cells expressing the minor alleles of two of the five TERT variants (A279T, {Delta}E441) had replication-dependent loss of telomere length. In an in vitro primer extension assay, these two variants showed reduced telomerase nucleotide addition processivity. Together, our data suggested that selective, common TERT variants could be revealed to harbour telomere maintenance defects, leading to a plausible explanation for their observed associations to telomere biology disorders.

genetics↗

bioRxiv2019

Temporal resolution of global gene expression and DNA methylation changes in the final phases of reprogramming towards induced pluripotency.

The generation of induced pluripotent stem cells (iPSCs) involves activation of the endogenous pluripotency circuitry and global DNA demethylation late in reprogramming, but temporal resolution of these events using existing markers is insufficient. Here, we generated murine transgenic lines harboring dual fluorescent reporters reflecting cell-state specific expression of the master pluripotency factor Oct4 and the 5-methylcytosine dioxygenase Tet1. By assessing reprogramming intermediates based on dual reporter patterns, we identified a sequential order of Tet1 and Oct4 gene activation at proximal and distal regulatory elements following pluripotency entry. Full induction of Tet1 marks a pivotal late intermediate stage occurring after a phase of global gene repression, and preceding full activation of Oct4 along with late naive pluripotency and germline-specific genes. Sequential activation of Tet1 further distinguishes two waves of global DNA demethylation, targeting distinct genomic features and largely uncoupled from transcriptional changes, with dynamics unique to iPSC reprogramming. Moreover, we demonstrate that loss of Tet1 is compatible with reprogramming towards full Oct4 gene activation, but generates iPSCs with aberrant DNA methylation, chromosomal instability during lineage priming and defective differentiation potential. Therefore, the transcriptional logic of Tet1 expression signals a deterministic epigenetic roadmap towards generation of high quality iPSCs.

cell biology↗