A5041944596- Single-cell and spatial transcriptomics
- Gene Regulatory Network Analysis
- Ferroelectric and Piezoelectric Materials
- Multiferroics and related materials
- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- HIV Research and Treatment
- Microwave Dielectric Ceramics Synthesis
- Interstitial Lung Diseases and Idiopathic Pulmonary Fibrosis
- Biochemical and Structural Characterization
- Machine Learning and Data Classification
- Congenital Diaphragmatic Hernia Studies
- Genomics and Phylogenetic Studies
- Tryptophan and brain disorders
- Advanced biosensing and bioanalysis techniques
- Machine Learning in Bioinformatics
- Neuroinflammation and Neurodegeneration Mechanisms
- Retinoids in leukemia and cellular processes
- Pluripotent Stem Cells Research
- Telomeres, Telomerase, and Senescence
- Plant Growth Enhancement Techniques
- Hepatitis C virus research
- Metabolomics and Mass Spectrometry Studies
- Explainable Artificial Intelligence (XAI)
- Biotin and Related Studies
Yale University
2021-2025
Nanjing University of Aeronautics and Astronautics
2023
Shanghai Jiao Tong University
2018-2019
Single-cell RNA-sequencing data has revolutionized our ability to understand of the patterns cell-cell and ligand-receptor connectivity that influence function tissues organs. However, quantification visualization these in a way informs tissue biology are major computational epistemological challenges. Here, we present Connectome, software package for R which facilitates rapid calculation interactive exploration signaling network topologies contained single-cell data. Connectome can be used...
Gram-negative bacteria use various secretion systems to deliver their secreted effectors. Among them, type IV system exists widely in a variety of bacterial species, and secretes effectors (T4SEs), which play vital roles host-pathogen interactions. However, experimental approaches identify T4SEs are time- resource-consuming. In the present study, we aim develop an silico stacked ensemble method predict whether protein is effector or not based on its sequence information. The sequences were...
Recent years have seen the release of several toolsets that reveal cell-cell interactions from single-cell data. However, all existing approaches leverage mean celltype gene expression values, and do not preserve fidelity original Here, we present NICHES (Niche Interactions Communication Heterogeneity in Extracellular Signaling), a tool to explore extracellular signaling at truly level.NICHES allows embedding ligand-receptor signal proxies visualize heterogeneous archetypes within cell...
Substance use disorders (SUD) and drug addiction are major threats to public health, impacting not only the millions of individuals struggling with SUD, but also surrounding families communities. One seminal challenges in treating studying human populations is high prevalence co-morbid conditions, including an increased risk contracting a immunodeficiency virus (HIV) infection. Of ~15 million people who inject drugs globally, 17% persons HIV. Conversely, HIV factor for SUD because chronic...
Abstract HIV infection exerts profound and long-lasting neurodegenerative effects on the central nervous system (CNS) that can persist despite antiretroviral therapy (ART). Here, we used single-nucleus multiome sequencing to map transcriptomic epigenetic landscapes of postmortem human brains from 13 healthy individuals 20 with who have a history treatment ART. Our study spanned three distinct regions—the prefrontal cortex, insular ventral striatum—enabling comprehensive exploration...
Abstract Principal component analysis (PCA) is indispensable for processing high-throughput omics datasets, as it can extract meaningful biological variability while minimizing the influence of noise. However, suitability PCA contingent on appropriate normalization and transformation count data, accurate selection number principal components; improper choices result in loss information or corruption signal due to excessive Typical approaches these challenges rely heuristics that lack...
Abstract Single-cell RNA-sequencing data can revolutionize our understanding of the patterns cell-cell and ligand-receptor connectivity that influence function tissues organs. However, quantification visualization these are major computational epistemological challenges. Here, we present Connectome , a software package for R which facilitates rapid calculation, interactive exploration, signaling network topologies contained in single-cell data. be used with any reference set known...
Abstract Summary Recent years have seen the release of several toolsets that reveal cell-cell interactions from single-cell data. However, all existing approaches leverage mean celltype gene expression values, and do not preserve fidelity original Here, we present NICHES ( N iche I nteractions C ommunication H eterogeneity in E xtracellular S ignaling), a tool to explore extracellular signaling at truly level. allows embedding ligand-receptor signal proxies visualize heterogeneous archetypes...
Tabular datasets with low-sample-size or many variables are prevalent in biomedicine. Practitioners this domain prefer linear tree-based models over neural networks since the latter harder to interpret and tend overfit when applied tabular datasets. To address these networks' shortcomings, we propose an intrinsically interpretable network for heterogeneous biomedical data. We design a locally sparse where local sparsity is learned identify subset of most relevant features each sample. This...
Bismuth ferrite–barium titanate (BFO–BT) is a well‐known lead‐free ferroelectric material with large electric field‐induced strain; however, its piezoelectricity very low under fields. Herein, by studying Ca‐doped BFO–BT ceramics, multiple structural evolutions are found in the domains at morphotropic phase boundary (MPB) BFO–BT‐based where part of local reveals transition from rhombohedral to tetragonal and relaxor phases. The superior electrical properties electrostrain S p 0.38%, P r 38.4...
Tissue homeostasis is controlled by cellular circuits governing cell growth, organization, and differentation. In this study we identify previously undescribed cell-to-cell communication that mediates information flow from mechanosensitive pleural mesothelial cells to alveolar-resident stem-like tuft in the lung. We find express a combination of mechanotransduction genes lineage-restricted ligands which makes them uniquely capable responding tissue tension producing paracrine cues acting on...
Background Arsenic has a broad anti‐cancer ability against hematologic malignancies and solid tumors. To systematically understand the biological functions of arsenic, we need to identify arsenic‐binding proteins in human cells. However, due lack effective theoretical tools experimental methods, only few have been identified. Methods Based on crystal structure ArsM, generated single mutation free energy profile for arsenic binding using perturbation methods. Multiple validations provide an...
Multi-modal high throughput biological data presents a great scientific opportunity and significant computational challenge. In multi-modal measurements, every sample is observed simultaneously by two or more sets of sensors. such settings, many variables in both modalities are often nuisance do not carry information about the phenomenon interest. Here, we propose unsupervised feature selection framework: identifying informative based on coupled high-dimensional measurements. Our method...
The (K,Na)NbO3(KNN)-based piezoelectric ceramics are one of the most promising lead-free materials to replace toxic lead-based ones for ultrasonic transducer applications owing their high Curie temperature and excellent properties. However, it is costly discover multiple doped compositions with enhanced properties based on traditional trial error approach. In this study, we proposed an efficient data-driven machine learning(ML) approach search KNN-based designed ML framework efficiently...
ABSTRACT During the pore-forming process, cholesterol-dependent cytolysins (CDCs) bind to cholesterol-rich membranes and subsequently undergo a series of conformational changes, predominantly involving in collapse protein transformation from α helices β-hairpins form large hydrophobic pore. In current study, we reconstructed structural model for both prepore complexes PFO based on cryo-EM data pneumolysin performed molecular dynamics simulations free energy calculations study changes...