Biology subjects

Dong, X.

Publications and source records attributed to Dong, X..

At least 325 records · Page 18Linked to original sources

bioRxiv2019

SCCNV: a software tool for identifying copy number variation from single-cell whole-genome sequencing

BackgroundIdentification of de novo mutations from cell populations requires single-cell whole-genome sequencing (SCWGS). Although many experimental protocols of SCWGS have been developed, few computational tools are available for downstream analysis of different types of somatic mutations, including copy number variation (CNV). ResultsWe developed SCCNV, a software tool for detecting CNVs from whole genome-amplified single cells. SCCNV is a read-depth based approach with adjustment for the whole-genome amplification bias. ConclusionsWe demonstrate its performance by analyzing data collected from most of the single-cell amplification methods, including DOP-PCR, MDA, MALBAC and LIANTI. SCCNV is freely available at https://github.com/biosinodx/SCCNV.

bioinformatics↗

bioRxiv2019

Single-cell whole-genome sequencing reveals the functional landscape of somatic mutations in B lymphocytes across the human lifespan

Introductory paragraphThe accumulation of mutations in somatic cells have been implicated as a cause of ageing since the 1950s1,2. Yet, attempts to establish a causal relationship between somatic mutations and ageing have been constrained by the lack of methods to directly identify mutational events in primary human tissues. Here we provide detailed, genome-wide mutation frequencies and spectra of human B lymphocytes from healthy individuals across the entire human lifespan, from newborns to centenarians, using a recently developed, highly accurate single-cell whole-genome sequencing method3. We found that the number of somatic mutations increases from <500 per cell in newborns to >3,000 per cell in centenarians. We discovered mutational hotspot regions, some of which, as expected, located at immunoglobulin genes associated with somatic hypermutation. B cell-specific mutation signatures were observed associated with development, ageing or somatic hypermutation (SHM). The SHM signature strongly correlated with the signature found in human chronic lymphocytic leukemia and malignant B-cell lymphomas4, indicating that even in B cells of healthy individuals the potential cancer-causing events are already present. We also identified multiple mutations in sequence features relevant to cellular function, i.e., transcribed genes and gene regulatory regions. Such mutations increased significantly during ageing, but only at approximately half the rate of the genome average, indicating selection against mutations that impact B cell function. This first full characterization of the landscape of somatic mutations in human B lymphocytes indicates that spontaneous somatic mutations accumulating with age can be deleterious and may contribute to both the increased risk for leukemia and the functional decline of B lymphocytes in the elderly.

genomics↗

bioRxiv2019

Spontaneous retrotranspositions in normal tissues are rare and associated with cell-type-specific differentiation

Activation of retrotransposons and their insertions into new genomic locations, i.e., retrotranspositions (RTs), have been identified in about 50% of tumors. However, the landscape of RTs in different, normal somatic cell types in humans remains largely unknown. Using single-cell whole-genome sequencing we identified 528 RT events, including LINE-1 (L1), and Alu, in 164 single cells and clones of fibroblasts, neurons, B lymphocytes, hepatocytes and liver stem cells, of 29 healthy human subjects aged from 0 to 106 years. The frequency of RTs was found to vary from <1 on average per cell in primary fibroblasts to 7.8 per cell in hepatocytes. Somewhat surprisingly, RT frequency does not increase with age, which is in contrast to other types of spontaneous mutation. RTs were found significantly more likely to insert in or close to target genes of the Polycomb Repressive Complex 2 (PRC2), which represses most of the genes encoding developmental regulators through H3K27me3 histone modification in embryonic stem cells. Indeed, when directly comparing RT frequency between differentiated liver hepatocytes with liver stem cells, the latter were almost devoid of RTs. These results indicate that spontaneous RTs are associated with cellular differentiation and occur, possibly, as a consequence of the transient chromatin transition of differentiation-specific genes from a transcriptionally repressed to activated state during the differentiation process.

genomics↗

bioRxiv2019

Structural basis for the recognition of K48-linked Ub chain by proteasomal receptor Rpn13

The interaction between K48-linked ubiquitin (Ub) chain and Rpn13 is important for proteasomal degradation of ubiquitinated substrate proteins. Only the complex structure between the N-terminal domain of Rpn13 (Rpn13NTD) and Ub monomer has been characterized, and it remains unclear how Rpn13 specifically recognizes K48-linked Ub chain. Using single-molecule FRET, here we show that K48-linked diubiquitin (K48-diUb) fluctuates among three distinct conformational states, and a preexisting compact state is selectively enriched by Rpn13NTD. The same binding mode is observed for full-length Rpn13 and longer K48-linked Ub chain. Using solution NMR spectroscopy, we have solved the complex structure between Rpn13NTD and K48-diUb. In the structure, Rpn13NTD simultaneously interacts with proximal and distal Ub subunits of K48-diUb that remain associated in the complex, thus corroborating smFRET findings. The proximal Ub interacts with Rpn13NTD similarly as the Ub monomer in the known Rpn13NTD:Ub structure, while the distal Ub binds to a largely electrostatic surface of Rpn13NTD. Thus, a charge reversal mutation in Rpn13NTD can weaken the interaction between Rpn13 and K48-linked Ub chain, causing accumulation of ubiquitinated proteins. Moreover, blockage of the access of the distal Ub to Rpn13NTD with a proximity attached Ub monomer can also disrupt the interaction between Rpn13 and K48-diUb. Together, the bivalent interaction of K48-linked Ub chain with Rpn13 provides the structural basis for Rpn13 linkage selectivity, which opens a new window for modulating proteasomal function.

biophysics↗

bioRxiv2019

High-quality genome assembly and high-density genetic map of asparagus bean

Asparagus bean (Vigna. unguiculata ssp. sesquipedialis), known for its very long and tender green pods, is an important vegetable crop broadly grown in the developing countries. Despite its agricultural and economic values, asparagus bean does not have a high-quality genome assembly for breeding novel agronomic traits. In this study, we reported a high-quality 632.8 Mb assembly of asparagus bean based on the whole genome shotgun sequencing strategy. We also generated a high-density linkage map for asparagus bean, which helped anchor 94.42% of the scaffolds into 11 pseudo-chromosomes. A total of 42,609 protein-coding genes and 3,579 non-protein-coding genes were predicted from the assembly. Taken together, these genomic resources of asparagus bean will facilitate the investigation of economically valuable traits in a variety of legume species, so that the cultivation of these plants would help combat the protein and energy malnutrition in the developing world.

genomics↗

bioRxiv2018

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

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

bioinformatics↗

bioRxiv2018

Skin Lesion Classification Via Combining Deep Learning Features and Clinical Criteria Representations

Skin lesion is a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of skin lesion is extremely challenging manually visualization. Hence, reliable automatic classification of skin lesions is meaningful to improve pathologists accuracy and efficiency. In this paper, we proposed a two-stage method to combine deep learning features and clinical criteria representations to address skin lesion automated diagnosis task. Stage 1 - feature encoding: Modified deep convolutional neural networks (CNNs, in this paper, we used Dense201 and Res50) were fine-tuned to extract rich image global features. To avoid hair noisy, we developed a lesion segmentation U-Net to mask out the decisive regions and used the masked image as CNNs inputs. In addition, color features, texture features and morphological features were exacted based on clinical criteria; Stage 2 - features fusion: LightGBM was used to select the salient features and model parameters, predicting diagnosis confidence for each category. The proposed deep learning frameworks were evaluated on the ISIC 2018 dataset. Experimental results show the promising accuracies of our frameworks were achieved.

bioinformatics↗

bioRxiv2018

Skin Lesion Segmentation with Improved C-UNet Networks

--This paper proposes an innovative method for Part1, skin lesion segmentation of the ISIC 2018 Challenge. Our network C-UNet is based on UNet network, we combined several methods on this basic network which made some improvements on Jaccard Index ultimately, our method yield an average Jaccard Index of 0.77 on the On-line validation dataset.

bioengineering↗

bioRxiv2018

Lesion Attributes Segmentation for Melanoma Detection with Deep Learning

Melanoma is the most deadly form of skin cancer worldwide. Many efforts have been made for early detection of melanoma. The International Skin Imaging Collaboration (ISIC) hosted the 2018 Challenges to help the diagnosis of melanoma based on dermoscopic images. In this paper, we describe our solutions for the task 2 of ISIC 2018 Challenges. We present two deep learning approaches to automatically detect lesion attributes of melanoma, one is a multi-task U-Net model and the other is a Mask R-CNN based model. Our multi-task U-Net model achieved a Jaccard index of 0.433 on official test data, which ranks the 5th place on the final leaderboard. The code for our solutions is publicly available.

bioinformatics↗

bioRxiv2018

Natural selection on TMPRSS6 associated with the blunted erythropoiesis and improved blood viscosity in Tibetan Pigs

Tibetan pigs, indigenous to Tibetan plateau, are well adapted to hypoxia. So far, there have been not any definitively described genes and functional sites responsible for hypoxia adaptation for the Tibetan pig. Here we conducted resequencing of the nearly entire genomic region (40.1 kb) of the candidate gene TMPRSS6 (Transmembrane protease, serine 6) associated with hemoglobin concentration (HGB) and red blood cell count (RBC) in 40 domestic pigs and 40 wild boars from five altitudes along the Tea-horse ancient road and identified 708 SNPs, in addition to an indel (CGTG/----) in the intron 10. Both the CGTG deletion frequency and the pairwise r2 linkage disequilibrium showed an increase with elevated altitudes in 838 domestic pigs from five altitudes, suggesting that TMPRSS6 has been under Darwinian positive selection. As the conserved core sequence of hypoxia-response elements (HREs), the deletion of CGTG in Tibetan pigs decreased the expression levels of TMPRSS6 mRNA and protein in the liver revealed by real-time quantitative PCR and western blot, respectively. To explore whether reduced TMPRSS6 expression level could improve blood viscosity, the relationship between CGTG indel and hematologic and hemorheologic parameters in 482 domestic pigs from continuous altitudes was detected and dissected a genetic effect on reducing HGB by 13.25g/L in Gongbogyamda Tibetan pigs and decreasing MCV by 4.79 fl in Diqing Tibetan pigs. In conclusion, the CGTG deletion of TMPRSS6 resulted in lower HGB and smaller MCV, thereby blunting erythropoiesis and improving blood viscosity as well as erythrocyte deformability.

physiology↗

bioRxiv2018

Development and Application of a High-Content Virion Display Human GPCR Array

G protein-coupled receptors (GPCRs) comprise the largest membrane protein family in humans and can respond to a wide variety of ligands and stimuli. Like other multi-pass membrane proteins, the biochemical properties of GPCRs are notoriously difficult to study because they must be embedded in lipid bilayers to maintain their native conformation and function. To enable an unbiased, high-throughput platform to profile biochemical activities of GPCRs in native conformation, we individually displayed 315 human non-odorant GPCRs (>85% coverage) in the envelope of human herpes simplex virus-1 and immobilized on glass to form a high-content Virion Display (VirD) array. Using this array, we found that 50% of the tested commercial anti-GPCR antibodies (mAbs) is ultra-specific, and that the vast majority of those VirD-GPCRs, which failed to be recognized by the commercial mAbs, could bind to their canonical ligands, indicating that they were folded correctly. Next, we used the VirD-GPCR arrays to examine binding specificity of two known peptide ligands and recovered expected interactions, as well as new off-target interactions, three of which were confirmed with real-time kinetics measurements. Finally, we explored the possibility of discovering novel pathogen targets by probing VirD-GPCR arrays with live group B Streptococcus (GBS), a common Gram-positive bacterium causing neonatal meningitis. Using cell invasion assays and a mouse model of hematogenous meningitis, we showed that inhibition of one of the five newly identified GPCRs, CysLTR1, greatly reduced GBS penetration in brain-derived endothelial cells and in mouse brains. Therefore, our work demonstrated that the VirD-GPCR array holds great potential for high-throughput, unbiased screening for small molecule drugs, affinity reagents, and deorphanization.

pharmacology and toxicology↗

bioRxiv2018

Fusion surface structure, function, and dynamics of gamete fusogen HAP2

HAP2 is a class II gamete fusogen in many eukaryotic kingdoms. A crystal structure of Chlamydomonas HAP2 shows a trimeric fusion state. Domains D1, D2.1 and D2.2 line the 3-fold axis; D3 and a stem pack against the outer surface. Surprisingly, hydrogen-deuterium exchange shows that surfaces of D1, D2.2 and D3 closest to the 3-fold axis are more dynamic than exposed surfaces. Three fusion helices in the fusion loop of each monomer expose hydrophobic residues at the trimer apex that are splayed from the 3-fold axis, leaving a solvent-filled cavity between the fusion loops in each monomer. At the base of the long fusion loop, Arg185 docks in a carbonyl cage. Comparisons to other structures, dynamics, and the greater effect on Chlamydomonas gamete fusion of mutation of axis-proximal than axis-distal fusion helices suggest that the apical portion of each monomer could tilt toward the 3-fold axis with merger of the fusion helices into a common fusion surface.

biochemistry↗

bioRxiv2018

BAP1 Loss Predicts Therapeutic Vulnerability in Malignant Peritoneal Mesothelioma

BackgroundMalignant Peritoneal Mesothelioma (PeM) is a rare but frequently fatal cancer that originates from the peritoneal lining of the abdomen. Standard treatment of PeM is limited to cytoreductive surgery and/or chemotherapy, and no effective targeted therapies for PeM yet exist. In the search for novel therapeutic target candidates in PeM, we performed a comprehensive integrative multi-omics analysis of 19 treatment-naive PeM tumors.\n\nResultsThe analysis identified PeM tumors with BAP1 loss to form a distinct molecular subtype characterized by distinct expression patterns of genes involved in chromatin remodeling, DNA repair pathway, and immune checkpoint receptor activation. This PeM subtype could potentially benefit from immune checkpoint, PARP, or HDAC inhibition therapies.\n\nConclusionsOur findings uncover BAP1 as a trackable prognostic and predictive biomarker, and refine PeM disease classification. This integrated molecular characterization provides a comprehensive foundation for developing PeM precision medicine.

cancer biology↗

bioRxiv2018

High-Speed Imaging of Paw Withdrawal Reflex to Objectively Assess Pain State in Mice

Rodents are often used for studying chronic pain mechanisms and developing new pain therapeutics, but objectively determining the animals pain state is a major challenge. To improve the precision of using reflexive withdrawal behaviors for interpreting the mouse pain state, we adopted high-speed videography to capture sub-second movement features of mice upon hind paw stimulation. We identified several parameters that are significantly different between behaviors evoked by innocuous and noxious stimuli, and combined them to map the mouse pain state through statistical modeling and machine learning. To test the utility of this approach, we determined the pain state triggered by von Frey hairs (VFHs) and optogenetic activation of two nociceptor populations. Our method reliably assesses the \"pain-like\" probability for each mouse paw withdrawal reflex under all scenarios, highlighting the improved precision of using this high resolution behavior-centered composite methodology to determine the mouse pain state from reflexive withdrawal assays.

neuroscience↗

bioRxiv2018

Genomes of trombidid mites reveal novel predicted allergens and laterally-transferred genes associated with secondary metabolism

BackgroundTrombidid mites have a unique lifecycle in which only the larval stage is ectoparasitic. In the superfamily Trombiculoidea (\"chiggers\"), the larvae feed preferentially on vertebrates, including humans. Species in the genus Leptotrombidium are vectors of a potentially fatal bacterial infection, scrub typhus, which affects 1 million people annually. Moreover, chiggers can cause pruritic dermatitis (trombiculiasis) in humans and domesticated animals. In the Trombidioidea (velvet mites), the larvae feed on other arthropods and are potential biological control agents for agricultural pests. Here, we present the first trombidid mites genomes, obtained both for a chigger, Leptotrombidium deliense, and for a velvet mite, Dinothrombium tinctorium.\n\nResultsSequencing was performed using Illumina technology. A 180 Mb draft assembly for D. tinctorium was generated from two paired-end and one mate-pair library using a single adult specimen. For L. deliense, a lower-coverage draft assembly (117 Mb) was obtained using pooled, engorged larvae with a single paired-end library. Remarkably, both genomes exhibited evidence of ancient lateral gene transfer from soil-derived bacteria or fungi. The transferred genes confer functions that are rare in animals, including terpene and carotenoid synthesis. Thirty-seven allergenic protein families were predicted in the L. deliense genome, of which nine were unique. Preliminary proteomic analyses identified several of these putative allergens in larvae.\n\nConclusionsTrombidid mite genomes appear to be more dynamic than those of other acariform mites. A priority for future research is to determine the biological function of terpene synthesis in this taxon and its potential for exploitation in disease control.

genomics↗

bioRxiv2018

A metaproteomics method to determine carbon sources and assimilation pathways of species in microbial communities

Measurements of the carbon stable isotope ratio ({delta}13C) are widely used in biology to address major questions regarding food sources and metabolic pathways used by organisms. Measurement of these so called stable carbon isotope fingerprints (SIFs) for microbes involved in biogeochemical cycling and microbiota of plants and animals have led to major discoveries in environmental microbiology. Currently, obtaining SIFs for microbial communities is challenging as the available methods either only provide limited taxonomic resolution, such as with the use of lipid biomarkers, or are limited in throughput, such as NanoSIMS imaging of single cells.\n\nHere we present \"direct Protein-SIF\" and the Calis-p software package (https://sourceforge.net/projects/calis-p/), which enable high-throughput measurements of accurate {delta}13C values for individual species within a microbial community. We benchmark the method using 20 pure culture microorganisms and show that the method reproducibly provides SIF values consistent with gold standard bulk measurements performed with an isotope ratio mass spectrometer. Using mock community samples, we show that SIF values can also be obtained for individual species within a microbial community. Finally, a case study of an obligate bacteria-animal symbiosis showed that direct Protein-SIF confirms previous physiological hypotheses and can provide unexpected new insights into the symbionts metabolism. This confirms the usefulness of this new approach to accurately determine {delta}13C values for different species in microbial community samples.\n\nSignificanceTo understand the roles that microorganisms play in diverse environments such as the open ocean and the human intestinal tract, we need an understanding of their metabolism and physiology. A variety of methods such as metagenomics and metaproteomics exist to assess the metabolism of environmental microorganisms based on gene content and gene expression. These methods often only provide indirect evidence for which substrates are used by a microorganism in a community. The direct Protein-SIF method that we developed allows linking microbial species in communities to the environmental carbon sources they consume by determining their stable carbon isotope signature. Direct Protein-SIF also allows assessing which carbon fixation pathway is used by autotrophic microorganisms that directly assimilate CO2.

microbiology↗

bioRxiv2017

De novo assemblies of high-quality reference transcriptomes identifies Rosaceae-common and Rosa-specific encoding genes

Roses are important plants for human beings with important economical and biological traits like continuous flowering, flower architecture, color and scent, that current model plants do not feature, thus are becoming ideal models for studying these traits. Due to high heterozygosity of rose genomes likely caused by frequent inter-species hybridization, a high-quality and well-annotated genome for Rosa plants is not available yet. Developing genetic and genomic tools with high quality has become necessary for further roses breeding and for disentangling the molecular genetic mechanisms underlying roses domestication. We here generated the high quality and comprehensive reference transcriptomes for Rosa chinensis Old Blush (OB) and R. wichuriana Basyes Thornless (BT), two roses contrasting at several important traits. These reference transcriptomes showed transcripts N50 above 2000bp. The two species shared about 23310 transcripts (N50 = 2364bp), among which about 8975 orthologs were conserved within genera of Rosa. Rosa plants shared about 5049 transcripts (Rosaceae-common) with these from Malus, Prunus, Rubus, and Fragaria. Finally, a pool of 417 transcripts unique to Rosa plants (Rosa-specific) was identified. These Rosaceae-common and Rosa-specific transcripts should facilitate the phylogenetic analysis of Rosaceae plants and investigation of Rosa-specific traits. The data reported here should provide the fundamental genomic tools and knowledge critical for understanding the biology and domestication of roses.

plant biology↗

bioRxiv2017

Prodomain-Growth Factor Swapping in the Structure of pro-TGF-β1

Transforming growth factor (TGF)-{beta} is synthesized as a proprotein that dimerizes in the endoplasmic reticulum. After processing in the Golgi to cleave the N-terminal prodomain from the C-terminal growth factor (GF) domain in each monomer, pro-TGF-{beta} is secreted and stored in latent complexes. It is unclear which prodomain and GF monomer are linked prior to proprotein convertase (PC) cleavage, and how much conformational change occurs following cleavage. We have determined a structure of pro-TGF-{beta}1 with the PC cleavage site mutated, to mimic the structure of the TGF-{beta}1 proprotein. Our structure demonstrates that the prodomain arm domain in one monomer is linked to the GF that interacts with the arm domain in the other monomer in the dimeric structure, i.e., the prodomain arm domain and GF domain in each monomer are swapped. Swapping has important implications for the mechanism of biosynthesis in the TGF-{beta} family and is relevant to the mechanism for preferential formation of heterodimers over homodimers for some members of the TGF-{beta} family. Our structure also provides comparisons between independent TGF-{beta}1 crystal structures and between human and porcine pro-TGF-{beta}1.

biochemistry↗