检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Liu, Yiling Dong, Juncheng Jiang, Ziyang Aloui, Ahmed Li, Keyu Klein, Hunter Tarokh, Vahid Carlson, David Program in Computational Biology and Bioinformatics Duke University School of Medicine DurhamNC United States Department of Electrical and Computer Engineering Duke University DurhamNC United States Department of Civil and Environmental Engineering Duke University DurhamNC United States Department of Biostatistics and Bioinformatics Duke University DurhamNC United States

Unsupervised domain adaptation (UDA) is a technique used to transfer knowledge from a labeled source domain to a different but related unlabeled target domain. While many UDA methods have shown success in the past, they often assume that the source and target domains must have identical class label distributions, which can limit their effectiveness in real-world scenarios. To address this limitation, we propose a novel generalization bound that reweights source classification error by aligning source and target sub-domains. We prove that our proposed generalization bound is at least as strong as existing bounds under realistic assumptions, and we empirically show that it is much stronger on real-world data. We then propose an algorithm to minimize this novel generalization bound. We demonstrate by numerical experiments that this approach improves performance in shifted class distribution scenarios compared to state-of-the-art methods. © 2023, CC BY.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Incorporating Prior Knowledge into Neural Networks through an Implicit Composite Kernel

arXiv

引用

arXiv 2022年

作者： Jiang, Ziyang Zheng, Tongshu Liu, Yiling Carlson, David Department of Civil and Environmental Engineering Duke University United States Division of Natural and Applied Science Duke Kunshan University China Program in Computational Biology and Bioinformatics Duke University School of Medicine United States Department of Civil and Environmental Engineering Department of Biostatistics and Bioinformatics Department of Computer Science Duke University United States

It is challenging to guide neural network (NN) learning with prior knowledge. In contrast, many known properties, such as spatial smoothness or seasonality, are straightforward to model by choosing an appropriate kernel in a Gaussian process (GP). Many deep learning applications could be enhanced by modeling such known properties. For example, convolutional neural networks (CNNs) are frequently used in remote sensing, which is subject to strong seasonal effects. We propose to blend the strengths of NNs and the clear modeling capabilities of GPs by using a composite kernel that combines a kernel implicitly defined by a neural network with a second kernel function chosen to model known properties (e.g., seasonality). We implement this idea by combining a deep network and an efficient mapping function based on either Nyström approximation or random Fourier features, which we call Implicit Composite Kernel (ICK). We then adopt a sample-then-optimize approach to approximate the full GP posterior distribution. We demonstrate that ICK has superior performance and flexibility on both synthetic and real-world datasets including a remote sensing dataset. The ICK framework can be used to include prior information into neural networks in many applications. © 2022, CC BY.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Decomposition of dynamic transcriptomic responses during effector-triggered immunity reveals conserved responses in two distinct plant cell populations

引用

Plant Communications 2024年第8期5卷 1-17页

作者： Xiaotong Liu Daisuke Igarashi Rachel A.Hillmer Thomas Stoddard You Lu Kenichi Tsuda Chad L.Myers Fumiaki Katagiri Department of Plant and Microbial Biology University of Minnesota-Twin CitiesSt PaulMN 55108USA Department of Computer Science and Engineering University of Minnesota-Twin CitiesMinneapolisMN 55455USA Bioinformatics and Computational Biology Graduate Program University of Minnesota-Twin CitiesMinneapolisMN 55455USA Institute for Innovation Ajinomoto Co.Inc.KawasakiJapan State Key Laboratory of Agricultural Microbiology Hubei Hongshan LaboratoryHubei Key Lab of Plant PathologyCollege of Plant Science and TechnologyHuazhong Agricultural UniversityWuhan 430070China Present address:225 S 6th St MinneapolisMN 55402USA Present address:Biotalys Inc.2520 Meridian Parkway Suite 480DurhamNC 27713USA Present address:Life Sciences Institute and Department of Cell and Developmental Biology University of Michigan Medical SchoolAnn ArborMI 48109USA

Rapid plant immune responses in the appropriate cells are needed for effective defense against *** transcriptome analysis is often used to describe overall immune responses,collection of transcriptome data with sufficient resolution in both space and time is *** reanalyzed public Arabidopsis time-course transcriptome data obtained after low-dose inoculation with a Pseudomonas syringae strain expressing the effector AvrRpt2,which induces effector-triggered immunity in ***-peak time-course patterns are prevalent among thousands of upregulated *** implemented a multicompartment modeling approach to decompose the double-peak pattern into two single-peak patterns for each *** decomposed peaks reveal an“echoing”pattern:the peak times of the first and second peaks correlate well across most upregulated *** demonstrated that the two peaks likely represent responses of two distinct cell populations that respond either cell autonomously or indirectly to ***,the peak decomposition has extracted spatial information from the time-course *** echoing pattern also indicates a conserved transcriptome response with different initiation times between the two cell populations despite different elicitor types.A gene set highly overlapping with the conserved gene set is also upregulated with similar kinetics during pattern-triggered *** of a WRKY network via different entry-point WRKYs can explain the similar but not identical transcriptome responses elicited by different elicitor *** discuss potential benefits of the properties of the WRKY activation network as an immune signaling network in light of pressure from rapidly evolving pathogens.

关键词： Arabidopsis Pseudomonas syringae plant immune response multi-compartment model WRKY transcription factors resilient immune signaling network

来源：评论

学校读者我要写书评

暂无评论

HiC-DC+ enables systematic 3D interaction calls and differential analysis for Hi-C and HiChIP

引用

Nature Communications 2021年第1期12卷 3366页

作者： Sahin, Merve Wong, Wilfred Zhan, Yingqian Van Deynze, Kinsey Koche, Richard Leslie, Christina S. Computational and Systems Biology Program Memorial Sloan Kettering Cancer Center New York NY United States Tri-Institutional Training Program in Computational Biology and Medicine New York NY United States Center for Epigenetics Research Memorial Sloan Kettering Cancer Center New York NY United States Bioinformatics Program University of California at San Diego San Diego CA United States

Recent genome-wide chromosome conformation capture assays such as Hi-C and HiChIP have vastly expanded the resolution and throughput with which we can study 3D genomic architecture and function. Here, we present HiC-DC+, a software tool for Hi-C/HiChIP interaction calling and differential analysis using an efficient implementation of the HiC-DC statistical framework. HiC-DC+ integrates with popular preprocessing and visualization tools and includes topologically associating domain (TAD) and A/B compartment callers. We found that HiC-DC+ can more accurately identify enhancer-promoter interactions in H3K27ac HiChIP, as validated by CRISPRi-FlowFISH experiments, compared to existing methods. Differential HiC-DC+ analyses of published HiChIP and Hi-C data sets in settings of cellular differentiation and cohesin perturbation systematically and quantitatively recovers biological findings, including enhancer hubs, TAD aggregation, and the relationship between promoter-enhancer loop dynamics and gene expression changes. HiC-DC+ therefore provides a principled statistical analysis tool to empower genome-wide studies of 3D chromatin architecture and function. © 2021, The Author(s).

关键词： Chromosomes Software Statistical methods

来源：评论

学校读者我要写书评

暂无评论

Entropic Optimal Transport Eigenmaps for Nonlinear Alignment and Joint Embedding of High-Dimensional Datasets

arXiv

引用

arXiv 2024年

作者： Landa, Boris Kluger, Yuval Ma, Rong Department of Electrical Engineering Yale University United States Program in Applied Mathematics Yale University United States Department of Biostatistics Harvard University United States Interdepartmental Program in Computational Biology and Bioinformatics Yale University United States Department of Pathology Yale University School of Medicine United States

Embedding high-dimensional data into a low-dimensional space is an indispensable component of data analysis. In numerous applications, it is necessary to align and jointly embed multiple datasets from different studies or experimental conditions. Such datasets may share underlying structures of interest but exhibit individual distortions, resulting in misaligned embeddings using traditional techniques. In this work, we propose Entropic Optimal Transport (EOT) eigenmaps, a principled approach for aligning and jointly embedding a pair of datasets with theoretical guarantees. Our approach leverages the leading singular vectors of the EOT plan matrix between two datasets to extract their shared underlying structure and align the datasets accordingly in a common embedding space. We interpret our approach as an inter-data variant of the classical Laplacian eigenmaps and diffusion maps embeddings, showing that it enjoys many favorable analogous properties. We then analyze a data-generative model where two observed high-dimensional datasets share latent variables on a common low-dimensional manifold, but each dataset is subject to data-specific translation, scaling, nuisance structures, and noise. We show that in a high-dimensional asymptotic regime, the EOT plan recovers the shared manifold structure by approximating a kernel function evaluated at the locations of the latent variables. Subsequently, we provide a geometric interpretation of our embedding by relating it to the eigenfunctions of population-level operators encoding the density and geometry of the shared manifold. Finally, we showcase the performance of our approach for data integration and embedding through simulations and analyses of real-world biological data, demonstrating its advantages over alternative methods in challenging scenarios. Copyright © 2024, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Origin of Overstretching Transitions in Single-Stranded Nucleic Acids

引用

Physical Review Letters 2013年第18期111卷 188302-188302页

作者： Zackary N. Scholl Mahir Rabbi David Lee Laura Manson Hanna S-Gracz Piotr E. Marszalek Program in Computational Biology and Bioinformatics Duke University Durham North Carolina 27708 USA Department of Mechanical Engineering and Materials Science Center for Biologically Inspired Materials and Material Systems Duke University Durham North Carolina 27708 USA Department of Molecular and Structural Biochemistry North Carolina State University Raleigh North Carolina 27708 USA

We combined single-molecule force spectroscopy with nuclear magnetic resonance measurements and molecular mechanics simulations to examine overstretching transitions in single-stranded nucleic acids. In single-stranded DNA and single-stranded RNA there is a low-force transition that involves unwinding of the helical structure, along with base unstacking. We determined that the high-force transition that occurs in polydeoxyadenylic acid single-stranded DNA is caused by the cooperative forced flipping of the dihedral angle formed between four atoms, O5’-C5’-C4’-C3’ (γ torsion), in the nucleic acid backbone within the canonical B-type helix. The γ torsion also flips under force in A-type helices, where the helix is shorter and wider as compared to the B-type helix, but this transition is less cooperative than in the B type and does not generate a high-force plateau in the force spectrums of A-type helices. We find that a similar high-force transition can be induced in polyadenylic acid single-stranded RNA by urea, presumably due to disrupting the intramolecular hydrogen bonding in the backbone. We hypothesize that a pronounced high-force transition observed for B-type helices of double stranded DNA also involves a cooperative flip of the γ torsion. These observations suggest new fundamental relationships between the canonical structures of single-and double-stranded DNA and the mechanism of their molecular elasticity.

关键词： NUCLEAR magnetic resonance spectroscopy SINGLE-stranded DNA NUCLEOTIDE sequence DEOXYRIBOSE ATOMIC force microscopy

来源：评论

学校读者我要写书评

暂无评论

Neural networks versus Logistic regression for 30 days all-cause readmission prediction

arXiv

引用

arXiv 2018年

作者： Allam, Ahmed Nagy, Mate Thoma, George Krauthammer, Michael Department of Quantitative Biomedicine University of Zurich Biomedical Informatics Group University Hospital of Zurich Program in Computational Biology and Bioinformatics Yale University Department of Pathology Yale School of Medicine Lister Hill National Center for Biomedical Communications National Library of Medicine

Heart failure (HF) is one of the leading causes of hospital admissions in the US. Readmission within 30 days after a HF hospitalization is both a recognized indicator for disease progression and a source of considerable financial burden to the healthcare system. Consequently, the identification of patients at risk for readmission is a key step in improving disease management and patient outcome. In this work, we used a large administrative claims dataset to (1) explore the systematic application of neural network-based models versus logistic regression for predicting 30 days all-cause readmission after discharge from a HF admission, and (2) to examine the additive value of patients’ hospitalization timelines on prediction performance. Based on data from 272,778 (49% female) patients with a mean (SD) age of 73 years (14) and 343,328 HF admissions (67% of total admissions), we trained and tested our predictive readmission models following a stratified 5-fold cross-validation scheme. Among the deep learning approaches, a recurrent neural network (RNN) combined with conditional random fields (CRF) model (RNNCRF) achieved the best performance in readmission prediction with 0.642 AUC (95% CI, 0.640-0.645). Other models, such as those based on RNN, convolutional neural networks and CRF alone had lower performance, with a non-timeline based model (MLP) performing worst. A competitive model based on logistic regression with LASSO achieved a performance of 0.643 AUC (95% CI, 0.640-0.646). We conclude that data from patient timelines improve 30 day readmission prediction for neural network-based models, that a logistic regression with LASSO has equal performance to the best neural network model and that the use of administrative data result in competitive performance compared to published approaches based on richer clinical datasets. Copyright © 2018, The Authors. All rights reserved.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Rapid Sampling of Hydrogen Bond Networks for computational Protein Design (vol 14, pg 2751, 2018)

引用

JOURNAL OF CHEMICAL THEORY AND COMPUTATION 2018年第10期14卷 5434-5434页

作者： Maguire, Jack B. Boyken, Scott E. Baker, David Kuhlman, Brian [†] Program in Bioinformatics and Computational Biology University of North Carolina at Chapel Hill Chapel Hill North Carolina 27599 United States [‡] Department of Biochemistry University of Washington Seattle Washington 98195 United States [§] Institute for Protein Design University of Washington Seattle Washington 98195 United States [∥] Howard Hughes Medical Institute University of Washington Seattle Washington 98195 United States [⊥] Department of Biochemistry and Biophysics University of North Carolina at Chapel Hill Chapel Hill North Carolina 27599 United States [#] Lineberger Comprehensive Cancer Center University of North Carolina at Chapel Hill Chapel Hill North Carolina 27599 United States

Hydrogen bond networks play a critical role in determining the stability and specificity of biomolecular complexes, and the ability to design such networks is important for engineering novel structures, interactions, and enzymes. One key feature of hydrogen bond networks that makes them difficult to rationally engineer is that they are highly cooperative and are not energetically favorable until the hydrogen bonding potential has been satisfied for all buried polar groups in the network. Existing computational methods for protein design are ill-equipped for creating these highly cooperative networks because they rely on energy functions and sampling strategies that are focused on pairwise interactions. To enable the design of complex hydrogen bond networks, we have developed a new sampling protocol in the molecular modeling program Rosetta that explicitly searches for sets of amino acid mutations that can form self-contained hydrogen bond networks. For a given set of designable residues, the protocol often identifies many alternative sets of mutations/networks, and we show that it can readily be applied to large sets of residues at protein-protein interfaces or in the interior of proteins. The protocol builds on a recently developed method in Rosetta for designing hydrogen bond networks that has been experimentally validated for small symmetric systems but was not extensible to many larger protein structures and complexes. The sampling protocol we describe here not only recapitulates previously validated designs with performance improvements but also yields viable hydrogen bond networks for cases where the previous method fails, such as the design of large, asymmetric interfaces relevant to engineering protein-based therapeutics.

关键词： Interfaces Peptides and proteins Monomers Noncovalent interactions Computer simulations

来源：评论

学校读者我要写书评

暂无评论

Integrative modeling of transmitted and de novo variants identifies novel risk genes for congenital heart disease

引用

Quantitative biology 2021年第2期9卷 216-227页

作者： Mo Li Xue Zeng Chentian Jin Sheng Chih Jin Weilai Dong Martina Brueckner Richard Lifton Qiongshi Lu Hongyu Zhao Department of Biostatistics Yale School of Public HealthNew HavenCT 06510USA Department of Genetics Yale UniversityNew HavenCT 06510USA Department of Molecular Cellular&Developmental BiologyYale UniversityCT 06510USA Department of Genetics Washington University School of MedicineSt LouisMO 63110USA Department of Pediatrics Yale UniversityNew HavenCT 06510USA Laboratory of Human Genetics and Genomics Rockefeller UniversityNew YorkNY 10065USA Department of Biostatistics and Medical Informatics University of Wisconsin-MadisonMadisonWl 53792USA Program of Computational Biology and Bioinformatics Yale UniversityNew HavenCT 06510USA

Background:Whole-exome sequencing(WES)studies have identified multiple genes enriched for de novo mutations(DNMs)in congenital heart disease(CHD)***,risk gene identification based on DNMs alone remains statistically challenging due to heterogenous etiology of CHD and low mutation rate in each ***:In this manuscript,we introduce a hierarchical Bayesian framework for gene-level association test which jointly analyzes de novo and rare transmitted *** integrative modeling of multiple types of genetic variants,gene-level annotations,and reference data from large population cohorts,our method accurately characterizes the expected frequencies of both de novo and transmitted variants and shows improved statistical power compared to analyses based on DNMs ***:Applied to WES data of 2,645 CHD proband-parent trios,our method identified 15 significant genes,half of which are novel,leading to new insights into the genetic bases of ***:These results showcase the power of integrative analysis of transmitted and de novo variants for disease gene discovery.

关键词： rare variants gene-level association test congenital heart disease de novo mutation

来源：评论

学校读者我要写书评

暂无评论

The subfornical organ is a nucleus for gut-derived T cells that regulate behaviour

引用

Nature 2025年 1-10页

作者： Tomomi M. Yoshida Mytien Nguyen Benjamin Y. Lu Di Xu Elizabeth Lin Daniel A. Waizman Mackenzie E. Coden Yifan Ma Kavita Israni-Winger Wenzhi Song Jafar Al Souz Joseph E. Craft Noah W. Palm David A. Hafler Andrew Wang Cuiling Zhang Le Zhang Katie N. Murray Anthony Russo Haowei Wang Jaime Grutzendler Biqing Zhu Hongyu Zhao Yann S. Mineur Joseph Luchsinger Marina R. Picciotto Sagar Bhatta Marcello Distasio Department of Immunobiology Yale School of Medicine New Haven CT USA Department of Rheumatology Allergy and Immunology Yale School of Medicine New Haven CT USA Department of Neurology Yale School of Medicine New Haven CT USA Department of Medicine (Section of Medical Oncology and Hematology) Yale School of Medicine New Haven CT USA Department of Neuroscience Yale School of Medicine New Haven CT USA Program of Computational Biology and Bioinformatics Yale University New Haven CT USA Department of Biostatistics School of Public Health Yale University New Haven CT USA Department of Psychiatry Yale School of Medicine New Haven CT USA Department of Pathology Yale School of Medicine New Haven CT USA Department of Ophthalmology and Visual Science Yale School of Medicine New Haven CT USA

Specialized immune cells that reside in tissues orchestrate diverse biological functions by communicating with parenchymal cells1. The contribution of the innate immune compartment in the meninges and the central nervous system (CNS) is well-characterized; however, whether cells of the adaptive immune system reside in the brain and are involved in maintaining homeostasis is unclear2,3,4. Here we show that the subfornical organ (SFO) of the brain is a nucleus for parenchymal αβ T cells in the steady-state brain in both mice and humans. Using unbiased transcriptomics, we show that these extravascular T cells in the brain are distinct from meningeal T cells: they secrete IFNγ robustly and express tissue-residence proteins such as CXCR6, which are required for their retention in the brain and for normal adaptive behaviour. These T cells are primed in the periphery by the microbiome, and traffic from the white adipose and gastrointestinal tissues to the brain. Once established, their numbers can be modulated by alterations to either the gut microbiota or the composition of adipose tissue. In summary, we find that CD4 T cells reside in the brain at steady state and are anatomically concentrated in the SFO in mice and humans; that they are transcriptionally and functionally distinct from meningeal T cells; and that they secrete IFNγ to maintain CNS homeostasis through homeostatic fat–brain and gut–brain axes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：