Biomedical subjects

Dong Xu

Publications and source records attributed to Dong Xu.

At least 19 recordsLinked to original sources

PubMed2025

Target and biomarker exploration portal for drug discovery.

MOTIVATION: The discovery of novel drug targets and precision biomarkers remains a major challenge in drug development, with traditional differential expression analysis often overlooking key regulatory proteins. Here, we present a novel, web-based bioinformatics tool, the Target and Biomarker Exploration Portal (TBEP), designed to accelerate the drug discovery process by integrating large-scale biomedical data with network analysis techniques. RESULTS: TBEP harnesses machine-learning approaches to mine and combine multimodal datasets, including human genetics, functional genomics, and protein-protein interaction networks, to decode causal disease mechanisms and uncover novel therapeutic targets and precision biomarkers for specific phenotypes. A unique feature of the tool is its ability to process large-scale data in real-time, facilitated by an efficient cloud-based architecture. Additionally, the tool incorporates an integrated large language model (LLM), which assists researchers in exploring and interpreting complex biological relationships within the generated networks and multi-omics data using natural language (English). By offering an intuitive, interactive interface, the LLM enhances the exploration of biological insights, making it easier for scientists to derive actionable conclusions. This powerful integration of network analysis, multi-omics data, and LLM provides a robust framework for accelerating the identification of novel drug targets. AVAILABILITY AND IMPLEMENTATION: The tool is publicly available at https://tbep.missouri.edu. The source code, documentation and installation instructions are available at GitHub repository: https://github.com/mizzoudbl/tbep.

Drug Discovery↗

PubMed2025

scBSP: a fast and accurate tool for identifying spatially variable features from high-resolution spatial omics data.

MOTIVATION: Emerging spatial omics technologies empower comprehensive exploration of biological systems from multi-omics perspectives in their native tissue location in 2D and 3D space. However, the limited sequencing depth, increasing spatial resolution, and growing spatial spots in spatial omics technologies present significant computational challenges in identifying biologically meaningful molecules with variable spatial distributions across various omics modalities. RESULTS: We introduce scBSP, an open-source, versatile, and user-friendly package for identifying spatially variable features in large-scale spatial omics data. scBSP demonstrates significantly enhanced computational efficiency, processing high-resolution spatial omics data within seconds, and exhibits robust cross-platform performance by consistently identifying spatially variable features with high reproducibility across various sequencing platforms. AVAILABILITY AND IMPLEMENTATION: scBSP is available for download from R CRAN at https://cran.r-project.org/web/packages/scBSP/index.html and PyPI at https://pypi.org/project/scbsp/.

Software↗

PubMed2025

Chromosome-level genome assembly of starry flounder (Platichthys stellatus).

Starry flounder (Platichthys stellatus) is widely distributed along the coastlines of the North Pacific. As an euryhaline flatfish, it can adapt to a wide range of environmental salinity ranging from freshwater to seawater, and is a promising aquaculture flatfish species in Korea and North China. However, no high-quality starry flounder reference genome has been reported to date, which greatly limits the studies of genetics and functional genomics. Here, we obtained a high-quality chromosome-level starry flounder genome assembly with a length of 643.56 Mb (scaffold N50: 26.19 Mb, contig N50: 10.00 Mb) combining short-reads sequencing, PacBio HiFi sequencing, and Hi-C sequencing. Approximately 94.02% of assembled sequences were anchored into 24 pseudochromosomes, and a total of 18 telomeres were detected. Totally 22,835 protein-coding genes and 227.87 Mb repetitive sequences were identified. In summary, the high-quality chromosome-level genome assembly not only provides valuable resources for genetic research in starry flounder, but also advances the development of molecular breeding technology of starry flounder.

Animals↗

PubMed2025

Optical genome mapping improves clinical interpretation of constitutional copy-number gains and reduces their VUS burden.

PURPOSE: Genomic structure of copy-number gains is critical for their clinical interpretation but cannot be determined by chromosomal microarray (CMA) analysis, which does not provide information about chromosomal location and orientation of multiplied regions. We thus hypothesized that in CMA testing gains have higher probability than losses to be classified as variants of uncertain significance (VUS) and that structural information from optical genome mapping (OGM) may improve their interpretation. METHODS: Using a χ2 test, we assessed the association between classification of copy-number variants as VUS and their type (gains vs losses) in a cohort of 4073 CMA cases. Thirty-three VUS gains involving disease-associated genes were characterized by OGM to evaluate if OGM data enable their more conclusive clinical interpretation. RESULTS: The proportion of variants reported as VUS compared with likely pathogenic/pathogenic was significantly higher for gains than losses, confirming their increased VUS burden. OGM successfully determined genomic structure for all 33 copy-number gains, showing that 26 of 33 were tandem duplications and 7 of 33 were complex rearrangements. Structural information facilitated clinical interpretation in majority of the cases; it supported benign nature for 27 of 33 gains and was inconclusive or supported pathogenic role for 6 of 33. An estimated 20% of reported VUS gains would not have been reportable if we had OGM data. CONCLUSION: We illustrate a specific advantage of OGM compared with CMA: in addition to detecting both copy-number variants and balanced rearrangements, OGM improves clinical interpretation of copy-number gains by providing structural information and is thus expected to significantly decrease their VUS burden.

Humans↗

PubMed2007

MUPRED: a tool for bridging the gap between template based methods and sequence profile based methods for protein secondary structure prediction.

Predicting secondary structures from a protein sequence is an important step for characterizing the structural properties of a protein. Existing methods for protein secondary structure prediction can be broadly classified into template based or sequence profile based methods. We propose a novel framework that bridges the gap between the two fundamentally different approaches. Our framework integrates the information from the fuzzy k-nearest neighbor algorithm and position-specific scoring matrices using a neural network. It combines the strengths of the two methods and has a better potential to use the information in both the sequence and structure databases than existing methods. We implemented the framework into a software system MUPRED. MUPRED has achieved three-state prediction accuracy (Q3) ranging from 79.2 to 80.14%, depending on which benchmark dataset is used. A higher Q3 can be achieved if a query protein has a significant sequence identity (>25%) to a template in PDB. MUPRED also estimates the prediction accuracy at the individual residue level more quantitatively than existing methods. The MUPRED web server and executables are freely available at http://digbio.missouri.edu/mupred.

Algorithms↗

PubMed2007

Graph embedding and extensions: a general framework for dimensionality reduction.

Over the past few decades, a large family of algorithms - supervised or unsupervised; stemming from statistics or geometry theory - has been designed to provide different solutions to the problem of dimensionality reduction. Despite the different motivations of these algorithms, we present in this paper a general formulation known as graph embedding to unify them within a common framework. In graph embedding, each algorithm can be considered as the direct graph embedding or its linear/kernel/tensor extension of a specific intrinsic graph that describes certain desired statistical or geometric properties of a data set, with constraints from scale normalization or a penalty graph that characterizes a statistical or geometric property that should be avoided. Furthermore, the graph embedding framework can be used as a general platform for developing new dimensionality reduction algorithms. By utilizing this framework as a tool, we propose a new supervised dimensionality reduction algorithm called Marginal Fisher Analysis in which the intrinsic graph characterizes the intraclass compactness and connects each data point with its neighboring points of the same class, while the penalty graph connects the marginal points and characterizes the interclass separability. We show that MFA effectively overcomes the limitations of the traditional Linear Discriminant Analysis algorithm due to data distribution assumptions and available projection directions. Real face recognition experiments show the superiority of our proposed MFA in comparison to LDA, also for corresponding kernel and tensor extensions.

Algorithms↗

PubMed2007

Cross-study analysis of the relative efficacies of oral antiviral therapies for chronic hepatitis B infection in nucleoside-naive patients.

BACKGROUND AND OBJECTIVE: Lamivudine and adefovir were approved for treatment of chronic hepatitis B virus (HBV) infection based on placebo-controlled trials, and entecavir was recently approved on the basis of its superiority over lamivudine in phase II/III trials; however, to date, these three therapies have not been compared head to head. METHODS: To evaluate the relative efficacy of these therapies, we applied a predefined protocol of established statistical techniques to compare data from phase III entecavir trials with published clinical trial results with lamivudine, adefovir and placebo in nucleoside-naive hepatitis B e antigen (HBeAg)-positive and -negative populations. RESULTS: A comprehensive literature search identified 612 publications/data sources, of which 28 satisfied predefined inclusion criteria. Independent reviewers extracted week 48-52 histological, virological, biochemical and serological endpoints from these sources, which were analysed with a fixed-effects model. For each of the three histological endpoints in HBeAg-positive patients (Histological Improvement, Ranked Assessment of Necroinflammation [RA-N] and Ranked Assessment of Fibrosis [RA-F]), entecavir was superior to adefovir. Entecavir was superior to lamivudine for Histological Improvement and comparable to lamivudine for RA-N and RA-F. With respect to reducing HBV DNA levels, entecavir (-6.98 log(10) copies/mL) was more effective than lamivudine (-5.46 log(10) copies/mL, p < 0.0001) and adefovir (-3.60 log(10) copies/mL, p < 0.0001), and lamivudine was more effective than adefovir (p < 0.0001). The parallel goals of HBV DNA reduction below the limit of quantitation (LOQ) [by polymerase chain reaction] and ALT normalisation were achieved more often with entecavir (69% and 67% of patients, respectively) than with lamivudine (38% and 59%, respectively; p < 0.0001 and p < 0.05, respectively) or adefovir (21% and 48%, respectively; both p < 0.0001), and more often with lamivudine than with adefovir (p < 0.0001 and p < 0.05, respectively). HBeAg seroconversion rates were higher with entecavir (21% of patients) and lamivudine (18%) than with adefovir (12%, p < 0.01 and p < 0.05, respectively). For each of the three histological endpoints in the HBeAg-negative population, entecavir was comparable to adefovir. Entecavir was superior to lamivudine for Histological Improvement, and comparable to lamivudine for RA-N and RA-F, and all three antivirals were superior to placebo. Entecavir proved superior to lamivudine and adefovir in lowering HBV DNA levels (-5.20 vs -4.66 vs -3.91 log(10) copies/mL, respectively; p < 0.0005 and p < 0.0001, respectively) and in suppressing HBV DNA below the LOQ (91% vs 73% vs 51% of patients, respectively; both p < 0.0001); in the latter respect, lamivudine was in turn superior to adefovir (p < 0.0001). Entecavir was also superior to lamivudine in normalising ALT (76% vs 69% patients, respectively; p < 0.05). CONCLUSIONS: Over a 12-month treatment period, this analysis predicts that the antiviral efficacy of entecavir would be superior to that of lamivudine, which in turn would be superior to that of adefovir, in nucleoside-naive patients with chronic HBV infection.

Adenine↗

PubMed2006

No supernovae associated with two long-duration gamma-ray bursts.

It is now accepted that long-duration gamma-ray bursts (GRBs) are produced during the collapse of a massive star. The standard 'collapsar' model predicts that a broad-lined and luminous type Ic core-collapse supernova accompanies every long-duration GRB. This association has been confirmed in observations of several nearby GRBs. Here we report that GRB 060505 (ref. 10) and GRB 060614 (ref. 11) were not accompanied by supernova emission down to limits hundreds of times fainter than the archetypal supernova SN 1998bw that accompanied GRB 980425, and fainter than any type Ic supernova ever observed. Multi-band observations of the early afterglows, as well as spectroscopy of the host galaxies, exclude the possibility of significant dust obscuration and show that the bursts originated in actively star-forming regions. The absence of a supernova to such deep limits is qualitatively different from all previous nearby long-duration GRBs and suggests a new phenomenological type of massive stellar death.

Journal Article↗

PubMed2006

A typical N-terminal extensions confer novel regulatory properties on GTP cyclohydrolase isoforms in Drosophila melanogaster.

The cofactor tetrahydrobiopterin plays critical roles in the modulation of the signaling molecules dopamine, serotonin, and nitric oxide. Deficits in cofactor synthesis have been associated with several human hereditary diseases. Responsibility for the regulation of cofactor pools resides with the first enzyme in its biosynthetic pathway, GTP cyclohydrolase I. Because organisms must be able to rapidly respond to environmental and developmental cues to adjust output of these signaling molecules, complex regulatory mechanisms are vital for signal modulation. Mammalian GTP cyclohydrolase is subject to end-product inhibition via an associated regulatory protein and to positive regulation via phosphorylation, although target residues are unknown. GTP cyclohydrolase is composed of a highly conserved homodecameric catalytic core and non-conserved N-terminal domains proposed to be regulatory sites. We demonstrate for the first time in any organism that the N-terminal arms of the protein serve regulatory functions. We identify two different modes of regulation of the enzyme mediated through the N-terminal domains. The first is end-product feedback inhibition, catalytically similar to that of the mammalian enzyme, except that feedback inhibition by the cofactor requires sequences in the N-terminal arms rather than a separate regulatory protein. The second is a novel inhibitory interaction between the N-terminal arms and the active sites, which can be alleviated through the phosphorylation of serine residues within the N termini. Both mechanisms allow for acute and highly responsive regulation of cofactor production as required by downstream signaling pathways.

Animals↗

PubMed2006

A fast SCOP fold classification system using content-based E-Predict algorithm.

BACKGROUND: Domain experts manually construct the Structural Classification of Protein (SCOP) database to categorize and compare protein structures. Even though using the SCOP database is believed to be more reliable than classification results from other methods, it is labor intensive. To mimic human classification processes, we develop an automatic SCOP fold classification system to assign possible known SCOP folds and recognize novel folds for newly-discovered proteins. RESULTS: With a sufficient amount of ground truth data, our system is able to assign the known folds for newly-discovered proteins in the latest SCOP v1.69 release with 92.17% accuracy. Our system also recognizes the novel folds with 89.27% accuracy using 10 fold cross validation. The average response time for proteins with 500 and 1409 amino acids to complete the classification process is 4.1 and 17.4 seconds, respectively. By comparison with several structural alignment algorithms, our approach outperforms previous methods on both the classification accuracy and efficiency. CONCLUSION: In this paper, we build an advanced, non-parametric classifier to accelerate the manual classification processes of SCOP. With satisfactory ground truth data from the SCOP database, our approach identifies relevant domain knowledge and yields reasonably accurate classifications. Our system is publicly accessible at http://ProteinDBS.rnet.missouri.edu/E-Predict.php.

Algorithms↗

PubMed2006

Inferring gene regulatory networks from multiple microarray datasets.

MOTIVATION: Microarray gene expression data has increasingly become the common data source that can provide insights into biological processes at a system-wide level. One of the major problems with microarrays is that a dataset consists of relatively few time points with respect to a large number of genes, which makes the problem of inferring gene regulatory network an ill-posed one. On the other hand, gene expression data generated by different groups worldwide are increasingly accumulated on many species and can be accessed from public databases or individual websites, although each experiment has only a limited number of time-points. RESULTS: This paper proposes a novel method to combine multiple time-course microarray datasets from different conditions for inferring gene regulatory networks. The proposed method is called GNR (Gene Network Reconstruction tool) which is based on linear programming and a decomposition procedure. The method theoretically ensures the derivation of the most consistent network structure with respect to all of the datasets, thereby not only significantly alleviating the problem of data scarcity but also remarkably improving the prediction reliability. We tested GNR using both simulated data and experimental data in yeast and Arabidopsis. The result demonstrates the effectiveness of GNR in terms of predicting new gene regulatory relationship in yeast and Arabidopsis. AVAILABILITY: The software is available from http://zhangorup.aporc.org/bioinfo/grninfer/, http://digbio.missouri.edu/grninfer/ and http://intelligent.eic.osaka-sandai.ac.jp or upon request from the authors.

Algorithms↗

PubMed2006

Mitochondrial aldehyde dehydrogenase attenuates hyperoxia-induced cell death through activation of ERK/MAPK and PI3K-Akt pathways in lung epithelial cells.

Oxygen toxicity is one of the major risk factors in the development of the chronic lung disease or bronchopulmonary dysplasia in premature infants. Using proteomic analysis, we discovered that mitochondrial aldehyde dehydrogenase (mtALDH or ALDH2) was downregulated in neonatal rat lung after hyperoxic exposure. To study the role of mtALDH in hyperoxic lung injury, we overexpressed mtALDH in human lung epithelial cells (A549) and found that mtALDH significantly reduced hyperoxia-induced cell death. Compared with control cells (Neo-A549), the necrotic cell death in mtALDH-overexpressing cells (mtALDH-A549) decreased from 25.3 to 6.5%, 50.5 to 9.1%, and 52.4 to 15.1% after 24-, 48-, and 72-h hyperoxic exposure, respectively. The levels of intracellular and mitochondria-derived reactive oxygen species (ROS) in mtALDH-A549 cells after hyperoxic exposure were significantly lowered compared with Neo-A549 cells. mtALDH overexpression significantly stimulated extracellular signal-regulated kinase (ERK) phosphorylation under normoxic and hyperoxic conditions. Inhibition of ERK phosphorylation partially eliminated the protective effect of mtALDH in hyperoxia-induced cell death, suggesting ERK activation by mtALDH conferred cellular resistance to hyperoxia. mtALDH overexpression augmented Akt phosphorylation and maintained the total Akt level in mtALDH-A549 cells under normoxic and hyperoxic conditions. Inhibition of phosphatidylinositol 3-kinase (PI3K) activation by LY294002 in mtALDH-A549 cells significantly increased necrotic cell death after hyperoxic exposure, indicating that PI3K-Akt activation by mtALDH played an important role in cell survival after hyperoxia. Taken together, these data demonstrate that mtALDH overexpression attenuates hyperoxia-induced cell death in lung epithelial cells through reduction of ROS, activation of ERK/MAPK, and PI3K-Akt cell survival signaling pathways.

Aldehyde Dehydrogenase↗

PubMed2006

Visualization of nitric oxide in living cells by a copper-based fluorescent probe.

Nitric oxide (NO) serves as a messenger for cellular signaling. To visualize NO in living cells, we synthesized a turn-on fluorescent probe for use in combination with microscopy. Unlike existing fluorescent sensors, the construct--a Cu(II) complex of a fluorescein modified with an appended metal-chelating ligand (FL)--directly and immediately images NO rather than a derivative reactive nitrogen species. Using spectroscopic and mass spectrometric methods, we established that the mechanism of the reaction responsible for the NO-induced fluorescence involves reduction of the complex to Cu(I) with release of the nitrosated ligand, which occurs irreversibly. We detected NO produced by both constitutive and inducible NO synthases (cNOS and iNOS, respectively) in live neurons and macrophages in a concentration- and time-dependent manner by using the Cu(II)-based imaging agent. Both the sensitivity to nanomolar concentrations of NO and the spatiotemporal information provided by this complex demonstrate its value for numerous biological applications.

Animals↗

PubMed2006

Down-regulation of P-glycoprotein expression in MDR breast cancer cell MCF-7/ADR by honokiol.

P-glycoprotein accounts for the most intrinsic and acquired cancer multidrug resistance. To inhibit the expression of P-glycoprotein is one of the effective ways to reverse cancer drug resistance. Honokiol, a naturally occurring compound, has been demonstrated to combat cancer through mechanisms including inhibition of angiogenesis and induction of apoptosis. Here, we show that honokiol down-regulated the expression of P-glycoprotein at mRNA and protein levels in MCF-7/ADR, a human breast MDR cancer cell line. The down-regulation of P-glycoprotein was accompanied with a partial recovery of the intracellular drug accumulation, and of the sensitivities toward adriamycin. This study reveals a novel function of honokiol as an anti-cancer agent.

ATP Binding Cassette Transporter, Subfamily B, Mem↗

PubMed2006

An RNA folding algorithm including pseudoknots based on dynamic weighted matching.

On the basis of maximum weighted matching (MWM) algorithm, we introduced a dynamic weight related with stem length and used a recursive algorithm to predict RNA secondary structures by searching the stem structure with maximum weight summation step-by-step. This algorithm not only avoids the complicated free energy calculation, but also it could attain higher prediction accuracy. Moreover, our algorithm can predict most types of potential pseudoknots in the RNA structure.

Algorithms↗

PubMed2006

Sequence diversity of the Bacillus thuringiensis and B. cereus sensu lato flagellin (H antigen) protein: comparison with H serotype diversity.

We set out to analyze the sequence diversity of the Bacillus thuringiensis flagellin (H antigen [Hag]) protein and compare it with H serotype diversity. Some other Bacillus cereus sensu lato species and strains were added for comparison. The internal sequences of the flagellin (hag) alleles from 80 Bacillus thuringiensis strains and 16 strains from the B. cereus sensu lato group were amplified and cloned, and their nucleotide sequences were determined and translated into amino acids. The flagellin allele nucleotide sequences for 10 additional strains were retrieved from GenBank for a total of 106 Bacillus species and strains used in this study. These included 82 B. thuringiensis strains from 67 H serotypes, 5 B. cereus strains, 3 Bacillus anthracis strains, 3 Bacillus mycoides strains, 11 Bacillus weihenstephanensis strains, 1 Bacillus halodurans strain, and 1 Bacillus subtilis strain. The first 111 and the last 66 amino acids were conserved. They were referred to as the C1 and C2 regions, respectively. The central region, however, was highly variable and is referred to as the V region. Two bootstrapped neighbor-joining trees were generated: a first one from the alignment of the translated amino acid sequences of the amplified internal sequences of the hag alleles and a second one from the alignment of the V region amino acid sequences, respectively. Of the eight clusters revealed in the tree inferred from the entire C1-V-C2 region amino acid sequences, seven were present in corresponding clusters in the tree inferred from the V region amino acid sequences. With regard to B. thuringiensis, in most cases, different serovars had different flagellin amino acid sequences, as might have been expected. Surprisingly, however, some different B. thuringiensis serovars shared identical flagellin amino acid sequences. Likewise, serovars from the same H serotypes were most often found clustered together, with exceptions. Indeed, some serovars from the same H serotype carried flagellins with sufficiently different amino acid sequences as to be located on distant clusters. Species-wise, B. halodurans, B. subtilis, and B. anthracis formed specific branches, whereas the other four species, all in the B. cereus sensu lato group, B. mycoides, B. weihenstephanensis, B. cereus, and B. thuringiensis, did not form four specific clusters as might have been expected. Rather, strains from any of these four species were placed side by side with strains from the other species. In the B. cereus sensu lato group, B. anthracis excepted, the distribution of strains was not species specific.

Amino Acid Sequence↗

PubMed2006

The ribosomal shunt translation strategy of cauliflower mosaic virus has evolved from ancient long terminal repeats.

We have screened portions of the large intergenic region of the Cauliflower mosaic virus (CaMV) genome for promoter activity in baker's yeast (Saccharomyces cerevisiae) and have identified an element that contributes to promoter activity in yeast but has negligible activity in plant cells when expressed in an agroinfiltration assay. A search of the yeast genome sequence revealed that the CaMV element had sequence similarity with the R region of the long terminal repeat (LTR) of the yeast Ty1 retrotransposon, with significant statistical confidence. In plants, the same CaMV sequence has been shown to have an essential role in the ribosomal shunt mechanism of translation, as it forms the base of the right arm of the stem-loop structure that is required for the ribosomal shunt. Since the left arm of the stem-loop structure must represent an imperfect reverse copy of the right arm, we propose that the ribosomal shunt has evolved from a pair of LTRs that have become incorporated end to end into the CaMV genome.

Base Sequence↗

PubMed2006

Rnall: an efficient algorithm for predicting RNA local secondary structural landscape in genomes.

BACKGROUND: The information of RNA local secondary structures (LSSs) can help retrieve biologically important motifs and study functions of RNA molecules. Most of the current RNA secondary structure prediction tools are not suitable for RNA LSS prediction on the genome scale due to high computational complexity. METHODS: We developed a new computer package Rnall based on a dynamic programming technique, which scans an RNA sequence with a sliding window and extracts all RNA LSSs with sizes no larger than the window size using the nearest neighbor thermodynamic parameters. The worst case running time of Rnall is O(W(3)L), where W is the window size and L is the query sequence length. In practice we observed a running time of O(W(2)L). We further introduced the concept of energy landscape for illustrating RNA LSS, which may facilitate RNA motif mining on the genomic scale. RESULTS: Rnall shows better prediction accuracy than two other prediction tools Lfold and Quickfold. Rnall is also applied to scan for RNA LSSs in three genomes, and the prediction maps well with known RNA motifs. CONCLUSIONS: Rnall is designed for RNA LSS prediction and together with the energy landscape, it has unique features that could be used for RNA structural motif mining. Rnall is freely available for download at http://digbio.missouri.edu/~wanx/Rnall or http://www.sysbio.muohio.edu/Rnall.

Algorithms↗