bioRxiv Science⌕ Search

Biology subjects

Ding, L.

Publications and source records attributed to Ding, L..

At least 145 records · Page 8Linked to original sources
bioRxiv2020

Single-nucleus RNA-Seq reveals a new type of brown adipocyte regulating thermogenesis

Adipose tissue usually is classified as either white, brown or beige/brite, based on whether it functions as an energy storage or thermogenic organ(Cannon and Nedergaard, 2004; Rosen and Spiegelman, 2014). It serves as an important regulator of systemic metabolism, exemplified by the fact that dysfunctional adipose tissue in obesity leads to a host of secondary metabolic complications such as diabetes, cardiovascular diseases and cancer(Hajer et al., 2008; Lauby-Secretan et al., 2016). In addition, adipose tissue is an important endocrine organ, which regulates the function of other metabolic tissues through paracrine and endocrine signals(Scheele and Wolfrum, 2019; Scherer, 2006). Work in recent years has demonstrated that tissue heterogeneity is an important factor regulating the functionality of various organs(Cao et al., 2017; Ginhoux et al., 2016; Park et al., 2018). Here we used single nucleus analysis in mice and men to deconvolute adipocyte heterogeneity. We are able to identify a novel subpopulation of adipocytes whose abundance is low in mice (2-8%) and which is increased under higher ambient temperatures. Interestingly, this population is abundant in humans who live close to thermoneutrality. We demonstrate that this novel adipocyte subtype functions as a paracrine cell regulating the activity of brown adipocytes through acetate-mediated regulation of thermogenesis. These findings could explain, why human brown adipose tissue is substantially less active than mouse tissue and targeting this pathway in humans might be utilized to restore thermogenic activity of this tissue.

cell biology↗
bioRxiv2020

EWS-FLI1 regulates and cooperates with core regulatory circuitry in Ewing sarcoma

Core regulatory circuitry (CRC)-dependent transcriptional network is critical for developmental tumors in children and young adults carrying few gene mutations. However, whether and how CRC contributes to transcription regulation in Ewing sarcoma is unknown. Here, we identify and functionally validate a CRC "trio" constituted by three transcription factors (TFs): KLF15, TCF4 and NKX2-2, in Ewing sarcoma cells. Epigenomic analyses demonstrate that EWS-FLI1, the primary fusion driver for this cancer, directly establishes super-enhancers of each of these three TFs to activate their transcription. In turn, KLF15, TCF4 and NKX2-2 co-bind to their own and each others super-enhancers and promoters, forming an inter-connected auto-regulatory loop. Functionally, CRC factors contribute significantly to cell proliferation of Ewing sarcoma both in vitro and in vivo, and are all overexpressed in this cancer. Mechanistically, CRC factors exhibit prominent capacity of co-regulating the epigenome in cooperation with EWS-FLI1, occupying 77.2% of promoters and 55.6% of enhancers genome-wide. Downstream, CRC TFs coordinately regulate gene expression networks in Ewing sarcoma, directly controlling important signaling pathways for cancer, such as lipid metabolism pathway, PI3K/AKT and MAPK signaling pathways. Together, molecular characterization of the oncogenic CRC model advances our understanding of the biology of Ewing sarcoma. Moreover, this study identifies CRC-downstream genes and signaling pathways, which may contain potential targets for therapeutic intervention for this malignancy.

cancer biology↗
bioRxiv2019

Effects of CXCL12/CXCR4/CXCR7 axis on human sperm motility and chemotaxis

Mammalian spermatozoa undergo a series of functional changes in the female genital tract before they can fertilize oocytes. Chemokines and their receptors are involved in these complex biological events. However, the detailed molecular mechanisms on the modulation of chemokines on spermatozoa still need to be elucidated. CXCL12, a chemokine, has been widely studied in leukocyte migration and. Here we found that CXCL12 was existent in human cumulus cells and follicular fluid and its receptors CXCR4 and CXCR7 were co-expressed on human spermatozoa. We investigated the effects of CXCL12 on various biological functions of human sperm and which receptor plays a dominant role in these processes. We found that CXCL12 could promote human sperm chemotaxis, motility, penetration in the mucus, acrosome reaction and Ca2+ influx through CXCR4 rather than CXCR7. In addition, the simplified physical model reasonably explained the change of sperm velocity under the influence of CXCL12 concentration gradient, which was identical to their physiological motion patterns.

cell biology↗
bioRxiv2019

Identifying Sialylation Linkages at the Glycopeptide Level by Glycosyltransferase Labeling Assisted Mass Spectrometry (GLAMS)

Precise assignment of sialylation linkages at the glycopeptide level is of importance in bottom-up glycoproteomics, and is also an indispensable step to understand the function of glycoproteins in pathogen-host interactions and cancer progression. Even though some efforts have been dedicated to the discrimination of 2,3/2,6-sialylated isomers, unambiguous identification of sialoglycopeptide isomers is still needed. Herein, an innovative strategy of glycosyltransferase labeling assisted mass spectrometry (GLAMS) was developed. After specific enzymatic labeling, oxonium ions from higher-energy C-trap dissociation (HCD) fragmentation of 2,3-sailoglycopeptides generate unique reporters to distinctly differentiate those of 2,6-sailoglycopeptide isomers. Using this strategy, a total of 1,236 linkage-specific sialoglycopeptides were successfully identified from 161 glycoproteins in human serum.\n\n\n\nO_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=129 SRC=\"FIGDIR/small/811554v2_ufig1.gif\" ALT=\"Figure 1\">\nView larger version (20K):\norg.highwire.dtl.DTLVardef@eada4borg.highwire.dtl.DTLVardef@a1a7bdorg.highwire.dtl.DTLVardef@10b72d9org.highwire.dtl.DTLVardef@a11ebd_HPS_FORMAT_FIGEXP M_FIG C_FIG

biochemistry↗
bioRxiv2019

Systematic Establishment of Robustness and Standards in Patient-Derived Xenograft Experiments and Analysis

Patient-Derived Xenografts (PDXs) are tumor-in-mouse models for cancer. PDX collections, such as those supported by the NCI PDXNet program, are powerful resources for preclinical therapeutic testing. However, variations in experimental design and analysis procedures have limited interpretability. To determine the robustness of PDX studies, the PDXNet tested temozolomide drug response for three pre-validated PDX models (sensitive, resistant, and intermediate) across four blinded PDX Development and Trial Centers (PDTCs) using independently selected SOPs. Each PDTC was able to correctly identify the sensitive, resistant, and intermediate models, and statistical evaluations were concordant across all groups. We also developed and benchmarked optimized PDX informatics pipelines, and these yielded robust assessments across xenograft biological replicates. These studies show that PDX drug responses and sequence results are reproducible across diverse experimental protocols. Here we share the range of experimental procedures that maintained robustness, as well as standardized cloud-based workflows for PDX exome-seq and RNA-Seq analysis and for evaluating growth.

cancer biology↗
bioRxiv2019

RecO impedes RecG-SSB binding to impair the strand annealing recombination pathway in E.coli

Faithful duplication of genomic DNA relies not only on the fidelity of DNA replication itself, but also on fully functional DNA repair and homologous recombination machinery. We report a molecular mechanism responsible for deciding homologous recombinational repair pathways during replication dictated by binding of RecO and RecG to SSB in E.coli. Using a RecG-yfp fusion protein, we found that RecG-yfp foci appeared only in the {Delta}recG, {Delta}recO and {Delta}recA, {Delta}recO double mutants. Surprisingly, foci were not observed in wild-type {Delta}recG, or double mutants where recG and either recF or, separately recR were deleted. In addition, formation of RecG-yfp foci in the {Delta}recO::kanR required wildtype ssb, as ssb-113 could not substitute. This suggests that RecG and RecO binding to SSB is competitive. We also found that the UV resistance of recO alone mutant increased to certain extent by supplementing RecG. In an ssb-113 mutant, RecO and RecG worked following a different pattern. Both RecO and RecG were able to participate in repairing UV damages when grown at permissive temperature, while they could also be involved in making DNA double strand breaks when grown at nonpermissive temperature. So, our results suggested that differential binding of RecG and RecO to SSB in a DNA replication fork in Escherichia coli.may be involved in determining whether the SDSA or DSBR pathway of homologous recombinational repair is used.\n\nAuthor summarySingle strand DNA binding proteins (SSB) stabilize DNA holoenzyme and prevent single strand DNA from folding into non-B DNA structures in a DNA replication fork. It has also been revealed that SSB can also act as a platform for some proteins working in DNA repair and recombination to access DNA molecules when DNA replication fork needs to be reestablished. In Escherichia coli, several proteins working primarily in DNA repair and recombination were found to participate in DNA replication fork resumption by physically interacting with SSB, including RecO and RecG etc. However the hierarchy of these proteins interacting with SSB in Escherichia coli has not been well defined. In this study, we demonstrated a differential binding of RecO and RecG to SSB in DNA replication was used to establish a RecO-dependent pathway of replication fork repair by abolishing a RecG-dependent replication fork repair. We also show that, RecG and RecO could randomly participate in DNA replication repair in the absence of a functional SSB, which may be responsible for the generation of DNA double strand breaks in an ssb-113 mutant in Escherichia coli.

molecular biology↗
bioRxiv2019

Depiction of secondary metabolites and antifungal activity of Bacillus velezensis DTU001

For a safe and sustainable environment, effective microbes as biocontrol agents are in high demand. We have isolated a new Bacillus velezensis strain DTU001, investigated its antifungal spectrum, sequenced its genome, and uncovered the production of lipopeptides in HPLC-HRMS analysis. To test the antifungal efficacy, extracts of B. velezensis DTU001 was tested against a range of twenty human or plant pathogenic fungi. We demonstrate that inhibitory potential of B. velezensis DTU001 against selected fungi is superior in comparison to single lipopeptide, either iturin or fengycin. The isolate showed analogous biofilm formation to other closely related Bacilli. To further support the biocontrol properties of the isolate, coculture with Candida albicans demonstrated that B. velezensis DTU001 exhibited excellent antiproliferation effect against C. albicans. In summary, the described isolate is a potential antifungal agent with a broad antifungal spectrum that might assist our aims to avoid hazardous pathogenic fungi and provide alternative to toxicity caused by chemicals.

microbiology↗
bioRxiv2019

The caudate nucleus contributes causally to decisions that combine noisy visual evidence and reward context

Our decisions often need to balance what we observe and what we desire. However, our understanding of how and where in the brain such decisions are made remains limited. A prime candidate for integrating sensory observations and desired rewards, and a focus of many modeling studies, is the basal ganglia pathway, which is known to make separate contributions to perceptual decisions that require the interpretation of uncertain sensory evidence and value-based decisions that select among outcome options 1-16. Here we report direct evidence for a causal role for a major input station of the basal ganglia, the caudate nucleus, in incorporating reward context and uncertain visual evidence to guide adaptive decision-making. In monkeys making saccadic decisions based on visual motion evidence and asymmetric reward-choice associations 17, single caudate neurons encoded information about both the visual evidence and the asymmetric rewards. Electrical microstimulation at caudate sites with task-modulated activity during motion viewing affected how the visual and reward information was used to form the decision. The microstimulation effects included coordinated changes in multiple computational components of the decision process, mimicking the monkeys voluntary adjustments in response to the asymmetric reward contexts. These results imply that the caudate nucleus plays key roles in coordinating the deliberative decision process that balances external evidence and internal preferences to guide adaptive behavior.

neuroscience↗
bioRxiv2019

Deep learning in bioinformatics: introduction, application, and perspective in big data era

Deep learning, which is especially formidable in handling big data, has achieved great success in various fields, including bioinformatics. With the advances of the big data era in biology, it is foreseeable that deep learning will become increasingly important in the field and will be incorporated in vast majorities of analysis pipelines. In this review, we provide both the exoteric introduction of deep learning, and concrete examples and implementations of its representative applications in bioinformatics. We start from the recent achievements of deep learning in the bioinformatics field, pointing out the problems which are suitable to use deep learning. After that, we introduce deep learning in an easy-to-understand fashion, from shallow neural networks to legendary convolutional neural networks, legendary recurrent neural networks, graph neural networks, generative adversarial networks, variational autoencoder, and the most recent state-of-the-art architectures. After that, we provide eight examples, covering five bioinformatics research directions and all the four kinds of data type, with the implementation written in Tensorflow and Keras. Finally, we discuss the common issues, such as overfitting and interpretability, that users will encounter when adopting deep learning methods and provide corresponding suggestions. The implementations are freely available at https://github.com/lykaust15/Deep_learning_examples.

bioinformatics↗
bioRxiv2019

Scavenging dicarbonyls with 5'-O-pentyl-pyridoxamine improves insulin sensitivity and reduces atherosclerosis through modulating inflammatory Ly6Chi monocytosis and macrophage polarization

ObjectiveOxidative stress contributes to the development of insulin resistance (IR) and atherosclerosis. Peroxidation of lipids produces reactive dicarbonyls such as Isolevuglandins (IsoLG) and malondialdehyde (MDA) that covalently bind plasma/cellular proteins, phospholipids, and DNA leading to altered function and toxicity. We examined whether scavenging reactive dicarbonyls with 5-O-pentyl-pyridoxamine (PPM) protects against the development of IR and atherosclerosis in Ldlr-/- mice. MethodsMale or female Ldlr-/- mice were fed a western diet (WD) for 16 weeks and treated with PPM versus vehicle alone. Plaque extent, dicarbonyl-lysyl adducts, efferocytosis, apoptosis, macrophage inflammation, and necrotic area were measured. Plasma MDA-LDL adducts and the in vivo and in vitro effects of PPM on the ability of HDL to reduce macrophage cholesterol were measured. Blood Ly6Chi monocytes and ex vivo 5-ethynyl-2-deoxyuridine (EdU) incorporation into bone marrow CD11b+ monocytes and CD34+ hematopoietic stem and progenitor cells (HSPC) were also examined. IR was examined by measuring fasting glucose/insulin levels and tolerance to insulin/glucose challenge. ResultsPPM reduced the proximal aortic atherosclerosis by 48% and by 46% in female and male Ldlr-/- mice, respectively. PPM also decreased IR and hepatic fat and inflammation in male Ldlr-/- mice. Importantly, PPM decreased plasma MDA-LDL adducts and prevented the accumulation of plaque MDA- and IsoLG-lysyl adducts in Ldlr-/- mice. In addition, PPM increased the net cholesterol efflux capacity of HDL from Ldlr-/- mice and prevented both the in vitro impairment of HDL net cholesterol efflux capacity and apoAI crosslinking by MPO generated hypochlorous acid. Moreover, PPM decreased features of plaque instability including decreased proinflammatory M1-like macrophages, IL-1{beta} expression, myeloperoxidase, apoptosis, and necrotic core. In contrast, PPM increased M2-like macrophages, Tregs, fibrous cap thickness, and efferocytosis. Furthermore, PPM reduced inflammatory monocytosis as evidenced by decreased blood Ly6Chi monocytes and proliferation of bone marrow monocytes and HSPC from Ldlr-/- mice. ConclusionsPPM has pleotropic atheroprotective effects in a murine model of familial hypercholesterolemia, supporting the therapeutic potential of reactive dicarbonyl scavenging in the treatment of IR and atherosclerotic cardiovascular disease.

pharmacology and toxicology↗
bioRxiv2019

Scavenging of reactive dicarbonyls with 2-hydroxybenzylamine reduces atherosclerosis in hypercholesterolemic Ldlr-/- mice

The pathogenesis of atherosclerosis may be accelerated by oxidative stress, which produces lipid peroxidation. Among the products of lipid peroxidation are highly reactive dicarbonyls including isolevuglandins (IsoLGs) and malondialdehyde (MDA) that covalently modify proteins. We investigated the impact of treatment with the dicarbonyl scavenger, 2-hydroxybenzylamine (2-HOBA) on HDL function and atherosclerosis in hyperlipidemic Ldlr-/- mice, a model of familial hypercholesterolemia (FH). Compared to mice treated with vehicle, 2-HOBA significantly decreased atherosclerosis in hypercholesterolemic Ldlr-/- mice by 31% in the proximal aortas and 60% in en face aortas, in the absence of changes in blood lipid levels. 2-HOBA reduced MDA content in HDL and LDL. Consuming a western diet increased plasma MDA-apoAI adduct levels in Ldlr-/- mice. 2-HOBA inhibited MDA-apoAI formation and increased the capacity of the mouse HDL to reduce macrophage cholesterol stores. Importantly, 2-HOBA reduced the MDA- and IsoLG-lysyl content in atherosclerotic aortas in Ldlr-/- mice. Furthermore, 2-HOBA diminished oxidative stress-induced inflammatory responses in macrophages, reduced the number of TUNEL-positive cells in atherosclerotic lesions by 72%, and decreased serum proinflammatory cytokines. Furthermore, 2-HOBA enhanced efferocytosis and promoted characteristics of stable plaque formation in mice as evidenced by a 69% (p<0.01) reduction in necrotic core and by increased collagen content (2.7-fold) and fibrous cap thickness (2.1-fold). HDL from patients with FH had increased MDA content resulting in a reduced ability of FH-HDL to decrease macrophage cholesterol content versus controls. Our results demonstrate that dicarbonyl scavenging with 2-HOBA has multiple atheroprotective effects on lipoproteins and reduces atherosclerosis in a murine model of FH, supporting its potential as a novel therapeutic approach for the prevention and treatment of human atherosclerotic cardiovascular disease.\n\nAbbreviations2-HOBA, 2-hydroxybenzylamine; 4-HOBA, 4-hydroxybenzylamine; MDA, malondialdehyde; 4-HNE, 4-hydroxynonenal; IsoLGs, isolevuglandins; HDL, high-density lipoproteins; LDL, lowdensity lipoprotein; LDLR, low-density lipoprotein receptor; ApoAI, apolipoprotein AI; ApoB, apolipoprotein B; ROS, reactive oxygen species; IL, interleukin.

pharmacology and toxicology↗
bioRxiv2019

Cryptococcus deuterogattii VGIIa infection associated with travel to the Pacific Northwest outbreak region in an anti-GM-CSF autoantibody positive patient in the United States

The Pacific Northwest (PNW), Vancouver Island, Oregon, and Washington have been the location of an ongoing Cryptococcus gattii outbreak since the 1990s, and there is evidence that the outbreak is expanding along the West Coast into California. Here we report a clinical case of a 69-year-old, HIV-negative man from North Carolina who was diagnosed with a fungal brain mass by magnetic resonance imaging (MRI) and pathology. He had traveled to Seattle and Vancouver three years earlier and Costa Rica four months prior to presentation. Phenotypic evidence shows the fungal mass isolated from the patients brain is C. gattii. In agreement with the phenotypic results, MLST provides genotypic evidence that assigns the infecting organism within in the C. gattii species complex and belonging to the C. deuterogattii VGIIa clade. Whole genome sequencing revealed >99.99% identity to the C. deuterogattii reference strain R265, indicating that the infecting strain is derived from the highly clonal outbreak strains in the PNW. We conclude the patient acquired the C. gattii infection during his travel to region three years prior and the infection was dormant for an extended period of time before causing disease. The patient tested positive for anti-granulocyte-macrophage colony-stimulating factor (GM-CSF) autoantibodies, supporting earlier reports that implicate these autoantibodies as a risk factor a risk factor associated with C. gattii infection.\n\nImportanceMortality rates associated with C. gattii infections are estimated to be between 13% and 33% depending on an individuals predisposition, and C. gattii has caused more at least 39 deaths in the PNW region. There have been four other international travel cases reported in patients from Europe and Asia with travel history to the PNW, but this study describes the first North American travel who acquired C. deuterogattii infection presenting within the United States, and the first case of a C. deuterogattii outbreak infection associated with anti-GM-CSF autoantibodies. Early and accurate diagnoses are important for disease prevention, treatment, and control of infectious diseases. Continual reporting of C. deuterogattii infections is necessary to raise awareness of the ongoing outbreak in the PNW and alert travelers and physicians to the endemic areas with potential risks.

microbiology↗
bioRxiv2019

RNAIndel: a machine-learning framework for discovery of somatic coding indels using tumor RNA-Seq data

Reliable identification of expressed somatic insertion/deletion (indels) is an unmet demand due to artifacts generated in PCR-based RNA-Seq library preparation and the lack of normal RNA-Seq data, presenting analytical challenges for discovery of somatic indels in tumor trasncriptome.\n\nBy implementing features characterized by PCR-free whole-genome and whole-exome sequencing into a machine-learning framework, we present RNAIndel, a tool for predicting somatic, germline and artifact indels from tumor RNA-Seq data alone. RNAIndel robustly predicts 87{square}93% of somatic indels from 235 samples with heterogeneous conditions, even recovering subclonal (VAF range 0.01-0.15) driver indels missed by targeted deep-sequencing, outperforming the current best-practice for RNA-Seq variant calling which had 57% sensitivity but with 12 times more false positives.\n\nRNAIndel is freely available at https://github.com/stjude/RNAIndel\n\nContactjinghui.zhang@stjude.org

bioinformatics↗
bioRxiv2018

The Allen Cell Structure Segmenter: a new open sourcetoolkit for segmenting 3D intracellular structures influorescence microscopy images

A continuing challenge in quantitative cell biology is the accurate and robust 3D segmentation of structures of interest from fluorescence microscopy images in an automated, reproducible, and widely accessible manner for subsequent interpretable data analysis. We describe the Allen Cell and Structure Segmenter (Segmenter), a Python-based open source toolkit developed for 3D segmentation of cells and intracellular structures in fluorescence microscope images. This toolkit brings together classic image segmentation and iterative deep learning workflows first to generate initial high-quality 3D intracellular structure segmentations and then to easily curate these results to generate the ground truths for building robust and accurate deep learning models. The toolkit takes advantage of the high-replicate 3D live cell image data collected at the Allen Institute for Cell Science of over 30 endogenous fluorescently tagged human induced pluripotent stem cell (hiPSC) lines. Each cell line represents a different intracellular structure with one or more distinct localization patterns within undifferentiated hiPS cells and hiPSC-derived cardiomyocytes. The Segmenter consists of two complementary elements, a classic image segmentation workflow with a restricted set of algorithms and parameters and an iterative deep learning segmentation workflow. We created a collection of 20 classic image segmentation workflows based on 20 distinct and representative intracellular structure localization patterns as a "lookup table" reference and starting point for users. The iterative deep learning workflow can take over when the classic segmentation workflow is insufficient. Two straightforward "human-in-the-loop" curation strategies convert a set of classic image segmentation workflow results into a set of 3D ground truth images for iterative model training without the need for manual painting in 3D. The deep learning model architectures used in this toolkit were designed and tested specifically for 3D fluorescence microscope images and implemented as readable scripts. The Segmenter thus leverages state of the art computer vision algorithms in an accessible way to facilitate their application by the experimental biology researcher. We include two useful applications to demonstrate how we used the classic image segmentation and iterative deep learning workflows to solve more challenging 3D segmentation tasks. First, we introduce the Training Assay approach, a new experimental-computational co-design concept to generate more biologically accurate segmentation ground truths. We combined the iterative deep learning workflow with three Training Assays to develop a robust, scalable cell and nuclear instance segmentation algorithm, which could achieve accurate target segmentation for over 98% of individual cells and over 80% of entire fields of view. Second, we demonstrate how to extend the lamin B1 segmentation model built from the iterative deep learning workflow to obtain more biologically accurate lamin B1 segmentation by utilizing multi-channel inputs and combining multiple ML models. The steps and workflows used to develop these algorithms are generalizable to other similar segmentation challenges. More information, including tutorials and code repositories, are available at allencell.org/segmenter.

cell biology↗
bioRxiv2018

LRScaf: Improving Draft Genomes Using Long Noisy Reads

BackgroundThe advent of Third Generation Sequencing (TGS) technologies opens the door to improve genome assembly. Long reads are promised to enhance the quality of fragmental draft assemblies constructed from Next Generation Sequencing (NGS) technologies. To date, a few of algorithms, i.e., SSPACE-LongRead, OPERA-LG, SMIS, npScarf, DBG2OLC, Unicycler, and LINKS, have been released that are capable of improving draft assemblies. However, hybrid assembly on large genomes is still challenging.\n\nResultsWe develop a scalable and computationally efficient scaffolder, Long Reads Scaffolder (LRScaf), that is capable of boosting assembly contiguity to a large extent using long reads. In our experiment, our method significantly improves the contiguity of human draft assemblies, increasing the NG50 value of CHM1 from 127.5 Kb to 10.4 Mb using 20-fold coverage PacBio dataset and the NG50 value of NA12878 from 115.7 Kb to 17.4 Mb using 35-fold coverage Nanopore dataset. The run time for the scaffolding procedure using LRScaf is the shortest in all cases of our experiment. Compared with the run time of SSPACE-LongRead, LRScaf is faster 300 times for S. cerevisiae and 2,300 times for D. melanogaster. The peak RAM of LRScaf, by contrast, is more efficient than LINKS in our test. For the rice case, the peak RAM of LINKS (877.72 Gb) is about 196 times higher than LRScaf. For the experiment of human assembly, the peak RAM of LINKS is beyond the capacity of system memory (1 Tb) whereas LRScaf takes 20.28 and 41.20 Gb on CHM1 and NA12878 datasets.\n\nConclusionsThe new method, LRScaf, yields the best or at least moderate contiguity and accuracy of scaffolds in the shortest run time compared with the state-of-the-art methods. Furthermore, it offers a new opportunity for the hybrid assembly of large genomes.

bioinformatics↗
bioRxiv2018

Concordance Of Genetic Variation That Increases Risk For Tourette Syndrome And That Influences Its Underlying Neurocircuitry

BACKGROUNDThere have been considerable recent advances in understanding the genetic architecture of Tourette Syndrome (TS) as well as its underlying neurocircuitry. However, the mechanisms by which genetic variations that increase risk for TS - and its main symptom dimensions - influence relevant brain regions are poorly understood. Here we undertook a genome-wide investigation of the overlap between TS genetic risk and genetic influences on the volume of specific subcortical brain structures that have been implicated in TS.\n\nMETHODSWe obtained summary statistics for the most recent TS genome-wide association study (GWAS) from the TS Psychiatric Genomics Consortium Working Group (4,644 cases and 8,695 controls) and GWAS of subcortical volumes from the ENIGMA consortium (30,717 individuals). We also undertook analyses using GWAS summary statistics of key symptom factors in TS, namely social disinhibition and symmetry behaviour. SNP Effect Concordance Analysis (SECA) was used to examine genetic pleiotropy - the same SNP affecting two traits - and concordance - the agreement in SNP effect directions across these two traits. In addition, a conditional false discovery rate (FDR) analysis was performed, conditioning the TS risk variants on each of the seven subcortical and the intracranial brain volume GWAS. Linkage Disequilibrium Score Regression (LDSR) was used as validation of SECA.\n\nRESULTSSECA revealed significant pleiotropy between TS and putaminal (p=2x10-4) and caudal (p=4x10-4) volumes, independent of direction of effect, and significant concordance between TS and lower thalamic volume (p=1x10-3). LDSR lent additional support for the association between TS and thalamic volume (p=5.85x10-2). Furthermore, SECA revealed significant evidence of concordance between the social disinhibition symptom dimension and lower thalamic volume (p=1x10-3), as well as concordance between symmetry behaviour and greater putaminal volume (p=7x10-4). Conditional FDR analysis further revealed novel variants significantly associated with TS (p<8x10-7) when conditioning on intracranial (rs2708146, q=0.046; and rs72853320, q=0.035 and hippocampal (rs1922786, q=0.001 volumes respectively.\n\nCONCLUSIONThese data indicate concordance for genetic variations involved in disorder risk and subcortical brain volumes in TS. Further work with larger samples is needed to fully delineate the genetic architecture of these disorders and their underlying neurocircuitry.

genomics↗
bioRxiv2018

SupportNet: a novel incremental learning framework through deep learning and support data

MotivationIn most biological data sets, the amount of data is regularly growing and the number of classes is continuously increasing. To deal with the new data from the new classes, one approach is to train a classification model, e.g., a deep learning model, from scratch based on both old and new data. This approach is highly computationally costly and the extracted features are likely very different from the ones extracted by the model trained on the old data alone, which leads to poor model robustness. Another approach is to fine tune the trained model from the old data on the new data. However, this approach often does not have the ability to learn new knowledge without forgetting the previously learned knowledge, which is known as the catastrophic forgetting problem. To our knowledge, this problem has not been studied in the field of bioinformatics despite its existence in many bioinformatic problems.\n\nResultsHere we propose a novel method, SupportNet, to solve the catastrophic forgetting problem efficiently and effectively. SupportNet combines the strength of deep learning and support vector machine (SVM), where SVM is used to identify the support data from the old data, which are fed to the deep learning model together with the new data for further training so that the model can review the essential information of the old data when learning the new information. Two powerful consolidation regularizers are applied to ensure the robustness of the learned model. Comprehensive experiments on various tasks, including enzyme function prediction, subcellular structure classification and breast tumor classification, show that SupportNet drastically outperforms the state-of-the-art incremental learning methods and reaches similar performance as the deep learning model trained from scratch on both old and new data.\n\nAvailabilityOur program is accessible at: https://github.com/lykaust15/SupportNet.

bioinformatics↗
bioRxiv2018

EZH2 co-opts gain-of-function p53 mutants to promote cancer growth and metastasis

With the unfolding of more and more cancer-driven gain-of-function (GOF) mutants of p53, it is important to define a common mechanism to systematically target different mutants rather than develop strategies tailored to inhibit each mutant individually. Here, using RNA immunoprecipitation sequencing (RIP-seq) we identified EZH2 as a p53 mRNA-binding protein. EZH2 bound to the internal ribosome entry site (IRES) in the 5 untranslated region (5UTR) of p53 mRNA and enhanced p53 protein translation in a methyltransferase-independent manner. EZH2 augmented p53 GOF mutant-mediated cancer growth and metastasis by increasing p53 GOF mutant protein level. EZH2 overexpression associated with the worse outcome only in patients with p53-mutated cancer. Depletion of EZH2 by antisense oligonucleotides inhibited p53 GOF mutant-mediated cancer growth. Our findings reveal a non-methyltransferase function of EZH2 that controls protein translation of p53 GOF mutants, inhibition of which causes synthetic lethality in cancer cells expressing p53 GOF mutants.

cancer biology↗