bioRxiv Science⌕ Search

Biology subjects

Shi, X.

Publications and source records attributed to Shi, X..

At least 235 records · Page 13Linked to original sources
bioRxiv2019

Functional analysis of KIT gene structural mutations causing porcine dominant white phenotype by using genome edited mouse models

Dominant white phenotype in pigs is considered to be caused by two structural mutations in KIT gene, including a 450-kb duplication encompassing the entire KIT gene, and a splice mutation (G > A) at the first base in intron 17, which leads to the deletion of exon 17 in mature KIT mRNA, and the production of KIT protein lacking a critical catalytic domain of kinase. However, this speculation has not yet been validated by functional studies. Here, by using CRISPR/Cas9 technology, we created two mouse models mimicing the structural mutations of KIT gene in dominant white pigs, including the splice mutation mouse model KIT D17/+ with exon 17 of one allele of KIT gene deleted, and duplication mutation mouse model KIT Dup/+ with one allele of KIT gene coding sequence (CDS) duplicated. We found that each mutation individually can not cause dominant white phenotype. Splice mutation homozygote is lethal and heterozygous mice present piebald coat. Inconsistent with previous speculation, we found KIT gene duplication mutation did not confer the patched phenotype, and had no obvious impact on coat color. Interestingly, combination of these two mutations lead to dominant white phenotype. Further molecular analysis revealed that combination of these two structural mutations could inhibit the kinase activity of the KIT protein, thus reduce the phosphorylation level of PI3K and MAPK pathway associated proteins, which may be related to the observed impaired migration of melanoblasts during embryonic development, and eventually lead to dominant white phenotype. Our study provides a further insight into the underlying genetic mechanisms of porcine dominant white coat colour.\n\nAuthor summaryKIT plays a critical role in control of coat colour in mammals. Two mutation coexistence in KIT are considered to be the cause of the Dominant white phenotype in pigs. One mutation is a 450-kb large duplication encompassing the entire KIT gene, another mutation is a splice mutation causing the skipping of KIT exon 17. The mechanism of these two mutations of KIT on coat color formation has not yet been validated. In this study, by using genome edited mouse models, we found each structural mutation individual does not lead dominant white phenotype, but combination of these two mutations could lead to a nearly complete white coat colour similar to pig dominant white phenotype, possibly due to the inhibition of the kinase activity of the KIT protein, thus its signalling function on PI3K and MAPK pathways, leading to impaired migration of melanoblasts during embryonic development, and eventually lead to dominant white phenotype. Our study provides a further insight into the underlying genetic mechanisms of porcine dominant white coat colour.

molecular biology↗
bioRxiv2019

MR-LDP: a two-sample Mendelian randomization for GWAS summary statistics accounting linkage disequilibrium and horizontal pleiotropy

The proliferation of genome-wide association studies (GWAS) has prompted the use of two-sample Mendelian randomization (MR) with genetic variants as instrumental variables (IV) for drawing reliable causal relationships between health risk factors and disease outcomes. However, the unique features of GWAS demand that MR methods account for both linkage disequilibrium (LD) and ubiquitously existing horizontal pleiotropy among complex traits, which is the phenomenon wherein a variant affects the outcome through mechanisms other than exclusively through the exposure. Therefore, statistical methods that fail to consider LD and horizontal pleiotropy can lead to biased estimates and false-positive causal relationships. To overcome these limitations, we propose a probabilistic model for MR analysis to identify the casual effects between risk factors and disease outcomes using GWAS summary statistics in the presence of LD and to properly account for horizontal pleiotropy among genetic variants (MR-LDP). MR-LDP utilizes a computationally efficient parameter-expanded variational Bayes expectation-maximization (PX-VBEM) algorithm to estimate the parameter of interest and further calibrates the evidence lower bound (ELBO) for a likelihood ratio test. We then conducted comprehensive simulation studies to demonstrate the advantages of MR-LDP over the existing methods in terms of both type-I error control and point estimates. Moreover, we used two real exposure-outcome pairs (CAD-CAD and Height-Height; CAD for coronary artery disease) to validate the results from MR-LDP compared with alternative methods, showing that our method is more efficient in using all instrumental variants in LD. By further applying MR-LDP to lipid traits and body mass index (BMI) as risk factors for complex diseases, we identified multiple pairs of significant causal relationships, including a protective effect of high-density lipoprotein cholesterol (HDL-C) on peripheral vascular disease (PVD), and a positive causal effect of body mass index (BMI) on hemorrhoids.

genetics↗
bioRxiv2019

Development of a potent and protective germline-like antibody lineage against Zika virus in a convalescent human

Zika virus (ZIKV) specific neutralizing antibodies hold a great promise for antibody-based interventions and vaccine design against ZIKV infection. However, their development in infected patients remain unknown. Here, we report on the dynamic development of a potent and protective ZIKV-specific human antibody ZK2B10 initially isolated from a ZIKV convalescent individual using next-generation sequencing (NGS). The unbiased repertoire analysis showed dramatic changes in many families of heavy and light chain variable regions. However, lineage tracing of ZK2B10 revealed limited somatic hypermutation throughout the 12 months since the onset of symptom. In particular, NGS-derived germline-like somatic variants neutralized and protected mice from lethal challenge of ZIKV without detectable cross-reactivity with Dengue virus (DENV). Site-directed mutagenesis identified two residues within {lambda} chain, N31 and S91 that are essential to the functional maturation. The dynamic features unveiled here will assist us to better understand the pathogenesis of ZIKV infection and inform rational design of vaccines.\n\nAuthor summaryRecently emerged ZIKV is associated with severe neurological complications such as Guillain-Barre syndrome in adults and congenital microcephaly in newborns. No ZIKV-specific therapeutics or vaccines are currently available. We and others have identified a number of neutralizing antibodies capable of protecting experimental animals from ZIKV infection. However, the development of these potent antibodies during ZIKV natural infection remains unknown. Here, we report on the longitudinal analysis of one such antibody ZK2B10 using next-generation sequencing (NGS), bioinformatics and functional analysis. We found that the ZK2B10 germline-like antibodies possess strong neutralizing activity in vitro and impressive protectivity against lethal ZIKV infection in vivo. These findings suggest that the potent and protective antibody response against ZIKV can be generated within relative short term with high germline identity which provide great hope and promise for successful vaccine development against ZIKV.

microbiology↗
bioRxiv2019

CoMM-S2: a collaborative mixed model using summary statistics in transcriptome-wide association studies

MotivationAlthough genome-wide association studies (GWAS) have deepened our understanding of the genetic architecture of complex traits, the mechanistic links that underlie how genetic variants cause complex traits remains elusive. To advance our understanding of the underlying mechanistic links, various consortia have collected a vast volume of genomic data that enable us to investigate the role that genetic variants play in gene expression regulation. Recently, a collaborative mixed model (CoMM) [42] was proposed to jointly interrogate genome on complex traits by integrating both the GWAS dataset and the expression quantitative trait loci (eQTL) dataset. Although CoMM is a powerful approach that leverages regulatory information while accounting for the uncertainty in using an eQTL dataset, it requires individual-level GWAS data and cannot fully make use of widely available GWAS summary statistics. Therefore, statistically efficient methods that leverages transcriptome information using only summary statistics information from GWAS data are required.\n\nResultsIn this study, we propose a novel probabilistic model, CoMM-S2, to examine the mechanistic role that genetic variants play, by using only GWAS summary statistics instead of individual-level GWAS data. Similar to CoMM which uses individual-level GWAS data, CoMM-S2 combines two models: the first model examines the relationship between gene expression and genotype, while the second model examines the relationship between the phenotype and the predicted gene expression from the first model. Distinct from CoMM, CoMM-S2 requires only GWAS summary statistics. Using both simulation studies and real data analysis, we demonstrate that even though CoMM-S2 utilizes GWAS summary statistics, it has comparable performance as CoMM, which uses individual-level GWAS data.\n\nContactjin.liu@duke-nus.edu.sg\n\nAvailability and implementationThe implement of CoMM-S2 is included in the CoMM package that can be downloaded from https://github.com/gordonliu810822/CoMM.\n\nSupplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗
bioRxiv2019

Interpretable attention model in transcription factor binding site prediction with deep neural networks

Due to the complexity of the biological factors that may influence the binding of transcription factors to DNA sequences, prediction of the potential binding sites remains a difficult task in computational biology. The attention mechanism in deep learning has shown its capability to learn from input features with long-range dependencies. Until now, no study has applied this mechanism in deep neural network models with input data from massively parallel sequencing. In this study, we aim to build a model for binding site prediction with the combination of attention mechanism and traditional deep learning techniques, including convolutional neural networks and recurrent neural networks. The performance of our methods is evaluated on the ENCODE-DREAM in vivo Transcription Factor Binding Site Prediction Challenge datasets.\n\nThe benchmark shows that our implementation with attention mechanism (called DeepGRN) improves the performance of the deep learning models. Our model achieves better performance in at least 9 of 13 targets than any of the methods participated in the DREAM challenge. Visualization of the attention weights extracted from the trained models reveals how those weights shift when binding signal peaks move along the genomic sequence, which can interpret how the predictions are made. Case studies show that the attention mechanism helps to extract useful features by focusing on regions that are critical to successful prediction while ignoring the irrelevant signals from the input.

bioinformatics↗
bioRxiv2019

ELIMAKI locus is required for mechanosensing and proprioception in birch trees

The remarkable vertical and radial growth observed in tree species, encompasses a major physical challenge for wood forming tissues. To compensate with increasing size and weight, cambium-derived radial growth increases the stem width, thereby supporting the aerial body of trees. This feedback appears to be part of a so-called \"proprioception\" (1, 2) mechanism that controls plant size and biomass allocation. Yet, how trees experience or respond to mechanical stress derived from their own vertical loading, remains unknown. Here, we combined two strategies to dissect the proprioceptive response in birch. First, we show that in response to physical loading, trees promote radial growth with different magnitudes along the stem. Next, we identified a mutant cultivar (B. pubescens cv. Elimaki) in which the main stem shows normal vertical development, but collapses after three months. By inducing precocious flowering, we generated a backcrossed population (BC1) by producing two generations in 4 years. In his scheme, we uncovered a recessive trait (eki) that segregates and genetically maps with a Mendelian monogenic pattern. Unlike WT, eki is resistant to vertical mechanical stimulation. However, eki responds normally to the gravitropic stimulus by making tension wood. Before the collapse, cell size in eki is compromised resulting in radial growth defects, depending on stem height. Cell walls of developing xylem and phloem tissues have delayed differentiation in eki, and its tissues are softer compared to WT as indicated by atomic force microscopy (AFM). The transcriptomic profile of eki highlighted the overlap with that of the Arabidopsis response to touch. Taken together, our results suggest that the mechanical environment and cell wall properties of developing woody tissues, can significantly affect the growth responses to vertical loading thereby compromising their proprioceptive capacity. Additionally, we introduce a fast forward genetics strategy to dissect complex phenotypes in trees.

plant biology↗
bioRxiv2019

Prediction of the Effect of Naturally Occurring Missense Mutations on Cellular N-Acetyl-Glucosaminidase Enzymatic Activity

In 2015, the Critical Assessment of Genome Interpretation (CAGI) proposed a challenge to devise a computational method for predicting the phenotypic consequences of genetic variants of a lysosomal hydrolase enzyme known as -N-acetylglucosaminidase (NAGLU). In 2014, the Human Gene Mutation Database released that 153 NAGLU mutations associated with MPS IIIB and 90 of them are missense mutations. The ExAC dataset catalogued 189 missense mutations NAGLU based on exome sequence data from about 60,000 individual and 24 of them are known to be disease associated. Biotechnology company, BioMarin, has quantified the relative functionality of NAGLU for the remaining subset of 165 missense mutations. For this particular challenge, we examined the subset missense mutations within the ExAC dataset and predicted the probability of a given mutation being deleterious and relating this measure to the capacity of enzymatic activity. In doing so, we hoped to learn the degree to which changes in amino acid physicochemical properties are tolerable for NAGLU function.\n\nAmino acid substitution (AAS) prediction methods are mainly based on the sequence and structure information. Simple comparisons between different AAS methods are not only difficult, but also irrational because each method was tested on various datasets and based on varied versions of databases. Currently, several AAS prediction methods have been introduced. PolyPhen-2, an updated version of PolyPhen, is a tool used to predict possible impacts of an amino acid substitution on the structure and function. Users are required to provide protein or SNP identifiers, protein sequences, substitution positions, etc. A score is provided, ranging from 0 to 1, corresponding to the probability of a mutation resulting in no functionality for the enzyme\n\nOnce the probability scores were generated, the dataset was then run through multiple machine learning algorithms to generate an applicable model for predicting the enzymatic activity of MPS IIIB-related mutations. This prediction was generated using the PolyPhen-2 probability score and other information about the mutation (amino acid type, location, allele frequency, etc.) as input feature variables. This generated a predicted aggregate score for each mutation, which was then reported back to CAGI. The results of the analysis are significant enough to hold confidence that the scores are decent predictors of enzymatic activity given a mutation in the NAGLU amino acid sequence.

bioinformatics↗
bioRxiv2019

Coloring Single Nanoparticle Trajectory on Live Cell Membrane with its Own History: a Presuppositionless Approach

Analyzing single particle trajectories is a prominent issue in understanding complex dynamics such as nanoparticle-cell interactions. Existing methods treat data points as isolated \"atoms\" and use predefined mechanical models to \"frame\" their complicated relationship. Herein, we propose a \"historical evolution\" based model-free strategy. It allows spatiotemporal heterogeneity embedded in a trajectory to self-emerge as consecutive colored segments before any model assumption, provide both an overall picture and local state transitions on the particle movement with minimum information loss, and inspire further model-based investigation. We demonstrate with simulations and experiments that the underlying mechanisms of various time-series and motion states of single nanoparticles on live cell membranes could all be revealed successfully. Since complexity studies at different levels of molecules, particles, cells, human beings, vehicles, and even stars could all be reduced to analyzing spatiotemporal trajectories of \"single particles\", this presuppositionless approach will help fundamental researches on many important systems.\n\nImpact StatementA preprocessing strategy for single particle trajectory analysis is established by providing an intuitive global pattern from \"historical experiences\" of the particle without predefining any mechanical models.

biophysics↗
bioRxiv2019

De novo Identification of Essential Protein Domains from CRISPR/Cas9 Tiling-sgRNA Knockout Screens

High-throughput CRISPR/Cas9 knockout screens using a tiling-sgRNA design permit in situ evaluation of protein domain function. To facilitate de novo identification of essential protein domains from such screens, we developed ProTiler, a computational method for the robust mapping of CRISPR knockout hyper-sensitive (CKHS) regions, which refers to the protein regions that are associated with strong sgRNA dropout effect in the screens. We used ProTiler to analyze a published CRISPR tiling screen dataset, and identified 175 CKHS regions in 83 proteins. Of these CKHS regions, more than 80% overlapped with annotated Pfam domains, including all of the 15 known drug targets in the dataset. ProTiler also revealed unannotated essential domains, including the N-terminus of the SWI/SNF subunit SMARCB1, which we validated experimentally. Surprisingly, the CKHS regions were negatively correlated with phosphorylation and acetylation sites, suggesting that protein domains and post-translational modification sites have distinct sensitivities to CRISPR/Cas9 mediated amino acids loss.

bioinformatics↗
bioRxiv2019

Identification of 67 pleiotropic genes for seven autoimmune diseases using multivariate statistical analysis

Although genome-wide association studies (GWAS) have a dramatic impact on susceptibility locus discovery, this univariate approach has limitation in detecting complex genotype-phenotype correlations. It is essential to identify shared genetic risk factors acting through common biological mechanisms of autoimmune diseases with a multivariate analysis. In this study, the GWAS summary statistics including 41,274 single nucleotide polymorphisms (SNPs) located in 11,516 gene regions was analyzed to identify shared variants of seven autoimmune diseases using metaCCA method. Gene-based association analysis was used to refine the pleiotropic genes. In addition, GO term enrichment analysis and protein-protein interaction network analysis were applied to explore the potential biological function of the identified genes. After metaCCA analysis, 4,962 SNPs (P<1.21x10-6) and 1,044 pleotropic genes (P<4.34x10-6) were identified. By screening the results of gene-based p-values, we identified the existence of 27 confirmed pleiotropic genes and highlighted 40 novel pleiotropic genes which achieved significance threshold in metaCCA analysis and were also associated with at least one autoimmune disease in the VEGAS2 analysis. The metaCCA method could identify novel variants associated with complex diseases incorporating different GWAS datasets. Our analysis may provide insights for some common therapeutic approaches of autoimmune diseases based on the pleiotropic genes and common mechanisms identified. Author summaryAlthough previous researches have clearly indicated varying degrees of overlapping genetic sensitivities in autoimmune diseases, it has proven GWAS only explain small percent of heritability. Here, we take advantage of recent technical and methodological advances to identify pleiotropic genes that act on common biological mechanisms and the overlapping pathophysiological pathways of autoimmune diseases. After selection using multivariate analysis and verification using gene-based analyses, we successfully identified a total of 67 pleiotropic genes and performed the functional term enrichment analysis. In particularly, 27 genes were identified to be pleiotropic in previous different types of studies, which were validated by our present study. Forty significant genes (16 genes were associated with one disease earlier, and 24 were novel) might be the novel pleiotropic candidate genes for seven autoimmune diseases. The improved detection not only yielded the shared genetic components but also provided better understanding for exploring the potential common biological pathogenesis of these major autoimmune diseases.

genetics↗
bioRxiv2019

Molecular subtyping and prognostic assessment based on tumor mutation burden in patients with lung adenocarcinomas

Tumor mutation burden (TMB) is a potential biomarker for response to immunotherapy. The subset of patients with TMB has not been well characterized in lung adenocarcinomas. Here we performed molecular subtyping based on TMB and compared the features of different subtypes including clinical features, somatic driver genes and mutational signatures. We found that patients with lower tumor mutation burden had a longer disease-free survival, while higher tumor mutation burden is associated with smoking and aging. Analysis of somatic driver genes and mutational signatures demonstrates a significant association between somatic RYR2 mutations and the subtype with higher mutation burden. Overall, our study identified two molecular subtypes based on TMB and described the corresponding difference in their clinical and genomic levels.

genomics↗
bioRxiv2019

Tumor Copy Number Deconvolution Integrating Bulk and Single-Cell Sequencing Data

Characterizing intratumor heterogeneity (ITH) is crucial to understanding cancer development, but it is hampered by limits of available data sources. Bulk DNA sequencing is the most common technology to assess ITH, but mixes many genetically distinct cells in each sample, which must then be computationally deconvolved. Single-cell sequencing (SCS) is a promising alternative, but its limitations -- e.g., high noise, difficulty scaling to large populations, technical artifacts, and large data sets -- have so far made it impractical for studying cohorts of sufficient size to identify statistically robust features of tumor evolution. We have developed strategies for deconvolution and tumor phylogenetics combining limited amounts of bulk and single-cell data to gain some advantages of single-cell resolution with much lower cost, with specific focus on deconvolving genomic copy number data. We developed a mixed membership model for clonal deconvolution via non-negative matrix factorization (NMF) balancing deconvolution quality with similarity to single-cell samples via an associated efficient coordinate descent algorithm. We then improve on that algorithm by integrating deconvolution with clonal phylogeny inference, using a mixed integer linear programming (MILP) model to incorporate a minimum evolution phylogenetic tree cost in the problem objective. We demonstrate the effectiveness of these methods on semi-simulated data of known ground truth, showing improved deconvolution accuracy relative to bulk data alone.

bioinformatics↗
bioRxiv2018

The genome of the fungal pathogen Verticillium dahliae reveals extensive bacterial to fungal gene transfer

Horizontal gene transfer (HGT) involves the transmission of genetic material between distinct evolutionary lineages and can be an important source of biological innovation. Reports of inter-kingdom HGT to eukaryotic microbial pathogens have accumulated over recent years. Verticillium dahliae is a notorious plant pathogen that causes vascular wilt disease on hundreds of plant species, resulting in high economic losses every year. Previously, the effector gene Ave1 and a glucosyltransferase-encoding gene were identified as virulence factor-encoding genes that were proposed to be horizontally acquired from a plant and a bacterial donor, respectively. However, to what extent HGT contributed to the overall genome composition of V. dahliae remained elusive. Here, we systematically searched for evidence of inter-kingdom HGT events in the genome of V. dahliae and provide evidence for extensive horizontal gene acquisition from bacterial origin.

microbiology↗
bioRxiv2018

In silico prediction and characterisation of secondary metabolite clusters in the plant pathogenic fungus Verticillium dahliae

Fungi are renowned producers of natural compounds, also known as secondary metabolites (SMs) that display a wide array of biological activities. Typically, the genes that are involved in the biosynthesis of SMs are located in close proximity to each other in so-called secondary metabolite clusters (SMCs). Many plant-pathogenic fungi secrete SMs during infection in order to promote disease establishment, for instance as cytocoxic compounds. Verticillium dahliae is a notorious plant pathogen that can infect over 200 host plants worldwide. However, the SM repertoire of this vascular pathogen remains mostly uncharted. To unravel the SM potential of V. dahliae, we performed in silico predictions and in-depth analyses of its SM clusters (SMC). We identified 25 potential SMCs in the V. dahliae genome, including loci that can be implicated in DHN-melanin, ferricrocin, triacetyl fusarinine and fujikurin production.

microbiology↗
bioRxiv2018

Deformed Alignment of Super-Resolution Images for Semi-flexible Structures in 3D

Due to low labeling efficiency and structural heterogeneity in fluorescence-based single-molecule localization microscopy (SMLM), image alignment and quantitative analysis is often required to make accurate conclusions on the spatial relationships between proteins. Cryo-electron microscopy (EM) image alignment procedures have been applied to average structures taken with super-resolution microscopy. However, unlike cryo-EM, the much larger cellular structures analyzed by super-resolution microscopy are often heterogeneous, resulting in misalignment. And the light-microscopy image library is much smaller, which makes classification not realistic. To overcome these two challenges, we developed a method to deform semi-flexible ring-shaped structures and then align the 3D structures without classification. These algorithms can register semi-flexible structures with an accuracy of several nanometers in short computation time and with greatly reduced memory requirements. We demonstrated our methods by aligning experimental Stochastic Optical Reconstruction Microscopy (STORM) images of ciliary distal appendages and simulated structures. Symmetries, dimensions, and locations of protein complexes in 3D are revealed by the alignment and averaging for heterogeneous, tilted, and under-labeled structures.

bioengineering↗
bioRxiv2018

Relative biological effect of alpha particle radiation on low dose phenomena: lethal mutation, hyper-radiosensitivity and increased radioresistance

At high doses, the current recommended radiation weighting factors advise a significantly higher effectiveness of alpha particles relative to gamma radiation. However, at lower doses, the ratio of effectiveness between radiations of varying linear energy transfer values is complicated due to the relative importance of low dose phenomena such as genomic instability, bystander effects, low dose hyper-radiosensitivity and increased radioresistance (HRS/IRR). Radium is the most common source of alpha radiation exposure to humans, but the dosimetry is complicated by the decay chain which involves gamma exposure due to radon daughters. This study aimed to isolate the relative biological effect of alpha particles after low doses of radium to cells and their progeny. This was done by subtracting the survival values of a human keratinocyte cell line (HaCaT) and an embryonic Chinook salmon cell line (CHSE-214) exposed to gamma irradiation, from survival of the same cell lines exposed to mixed alpha and gamma radiation through chronic exposure to Ra-226 and its decay products. The human cell line showed increased radioresistance when exposed to low doses of alpha particles. In contrast the fish cell line, which demonstrated radioresistance to low dose gamma energy, demonstrated increased lethality when exposed to low doses of alpha particles. The results confirm the need to consider the dose-response relationship when developing radiation weighting factors for low dose exposures, as well as the need to be aware of possible cell line and species differences.

cell biology↗
bioRxiv2018

Local Enrichment of HP1alpha at Telomeres Alters Their Structure and Regulation of Telomere Protection

Enhanced telomere maintenance is evident in malignant cancers. While telomeres are thought to be inherently heterochromatic, detailed mechanisms of how epigenetic modifications impact telomere protection and structures are largely unknown in human cancers. Here we develop a molecular tethering approach to experimentally enrich heterochromatin protein HP1 specifically at telomeres. This results in increased deposition of H3K9me3 at cancer cell telomeres. Telomere extension by telomerase is attenuated, and damage-induced foci at telomeres are reduced, indicating augmentation of telomere stability. Super resolution STORM imaging shows an unexpected increase in irregularity of telomeric structure. Telomere-tethered chromo shadow domain (CSD) mutant I165A of HP1 abrogates both the inhibition of telomere extension and the irregularity of telomeric structure, suggesting the involvement of at least one HP1-ligand in mediating these effects. This work presents a new approach to specifically manipulate the epigenetic status locally at telomeres to uncover insights into molecular mechanisms underlying telomere structural dynamics.

cell biology↗