检索结果-内蒙古大学图书馆

Beyond current boundaries: Integrating deep learning and AlphaFold for enhanced protein structure prediction from low-resolution cryo-EM maps

引用

Computational Biology and Chemistry 2025年 119卷 108494页

作者： Xin Ma Dong Si Division of Computing and Software Systems University of Washington Bothell 18115 Campus Way NE Bothell 98011 WA USA

Constructing atomic models from cryo-electron microscopy (cryo-EM) maps is a crucial yet intricate task in structural biology. While advancements in deep learning, such as convolutional neural networks (CNNs) and graph neural networks (GNNs), have spurred the development of sophisticated map-to-model tools like DeepTracer and ModelAngelo, their efficacy notably diminishes with low-resolution maps beyond 4 Å. To address this critical gap, this study introduces DeepTracer-LowResEnhance, an innovative computational framework that uniquely integrates structural predictions from AlphaFold with a deep-learning-based map refinement strategy specifically tailored to enhance low-resolution maps. Unlike existing techniques, our approach leverages the strengths of AlphaFold’s sequence-based predictions combined with advanced neural network-driven refinement processes to significantly improve map interpretability and modeling accuracy. DeepTracer-LowResEnhance demonstrates substantial and consistent improvements on an extensive dataset comprising 37 diverse protein cryo-EM maps, covering resolutions from 2.5 to 8.4 Å and including 22 challenging cases below 4 Å resolution. DeepTracer-LowResEnhance achieves an average TM-score improvement of 3.53x compared to baseline DeepTracer predictions. Notably, our enhanced methodology showed performance gains across 95.5% of the tested low-resolution datasets. A comparative analysis alongside traditional sharpening methods such as Phenix’s auto-sharpening illustrates DeepTracer-LowResEnhance’s superior capability in rendering more detailed and precise atomic models, thereby pushing the boundaries of current computational structural biology methodologies.

关键词： AlphaFold Computational structural biology Cryo-EM Deep learning

来源：评论

学校读者我要写书评

暂无评论

Using DragPushing to Refine Concept Index for Text Categorization

引用

Journal of Computer Science & Technology 2006年第4期21卷 592-596页

作者：程学旗谭松波 Lilian Tang Division of Intelligent Software Systems Institute of Computing Technology Chinese Academy of SciencesBeijing 100080 P.R. China Department of Computing University of Surrey U.K.

Concept index （CI） is a very fast and efficient feature extraction （FE） algorithm for text classification. The key approach in CI scheme is to express each document as a function of various concepts （centroids） present in the collection. However,the representative ability of centroids for categorizing corpus is often influenced by so-called model misfit caused by a number of factors in the FE process including feature selection to similarity measure. In order to address this issue, this work employs the ＂DragPushing＂ Strategy to refine the centroids that are used for concept index. We present an extensive experimental evaluation of refined concept index （RCI） on two English collections and one Chinese corpus using state-of-the-art Support Vector Machine （SVM） classifier. The results indicate that in each case, RCI-based SVM yields a much better performance than the normal CI-based SVM but lower computation cost during training and classification phases.

关键词： text classification information retrieval machine learning

来源：评论

学校读者我要写书评

暂无评论

Chinese Landmark Recognition

Chinese Landmark Recognition

引用

2020 International Conference on computing, Networking and Communications, ICNC 2020

作者： Wu, Hao Chen, Min University of Washington Bothell Division of Computing and Software Systems School of STEM United States

ISBN: (纸本)9781728149059

Landmark recognition is a useful yet challenging task due to the lack of large annotated landmark datasets. Previous related work often neglects non-English speaking areas, such as Chinese landmarks. Tests using commercial landmark recognition APIs also indicate the incompleteness of such landmark data. Our project releases Chinese landmark and attraction dataset for recognition of human-made and natural landmarks in China. The dataset contains 42,548 images depicting 987 unique landmarks. Moreover, we build a landmark recognition model based on Convolution Neural Network and transfer learning techniques. Lastly, an iOS application is built based on our landmark recognition model. It allows user to do landmark prediction on images taken by camera and images from device gallery. © 2020 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Interpretable Deep Learning Models With Concept Whitening Layers

Interpretable Deep Learning Models With Concept Whitening La...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Boddam, Megana Kim, Wooyoung University of Washington Bothell Division of Computing and Software Systems School of Stem BothellWA United States

ISBN: (纸本)9798350337488

Deep Learning models, such as convolutional neural networks (CNN), are hard to interpret due to their complex, nonlinear, and high-dimensional algorithms. We focused on interpreting CNNs for predicting Hepatocellular Carcinoma (HCC), the most common primary liver cancer, using histopathology images. We used a 50-layer residual neural network (RNN50), an inception CNN, and a recent innovation called concept whitening (CW). The results showed that adding CW layers to CNN models did not significantly affect model accuracy. However, post-hoc analysis methods, including inter-concept similarity rating, intra-concept similarity rating, concept importance rating, and feature vector displays, revealed improved interpretability. © 2023 IEEE.

关键词： Concept Vectors Concept Whitening Convolutional Neural Networks Deep Learning Feature Vectors Interpretive Machine Learning Residual Neural Networks

来源：评论

学校读者我要写书评

暂无评论

Maintaining stream statistics over multiscale sliding windows

Maintaining stream statistics over multiscale sliding window...

引用

作者： Jiao, Yishan Institute of Computing Technology Chinese Academy of Sciences Division of Intelligent Software Systems Institute of Computing Technology Chinese Academy of Sciences Zhongguancun Beijing China

In this article, we propose a new multiscale sliding window model which differentiates data items in different time periods of the data stream, based on a reasonable monotonicity of resolution assumption. Our model, as a well-motivated extension of the sliding window model, stands halfway between the traditional all-history and time-decaying models. We also present algorithms for estimating two significant data stream statistics - -F0 and Jacard's similarity coefficient - -with reasonable accuracies under the new model. Our algorithms use space logarithmic in the data stream size and linear in the number of windows;they support update time logarithmic in the number of windows and independent of the accuracy required. Our algorithms are easy to implement. Experimental results demonstrate the efficiencies of our algorithms. Our techniques apply to scenarios in which universe sampling is used. © 2006 ACM.

关键词： Database systems

来源：评论

学校读者我要写书评

暂无评论

BrainGrid+Workbench: High-performance/high-quality neural simulation

BrainGrid+Workbench: High-performance/high-quality neural si...

引用

2017 International Joint Conference on Neural Networks, IJCNN 2017

作者： Stiber, Michael Kawasaki, Fumitaka Davis, Delmar B. Asuncion, Hazeline U. Lee, Jewel Yun-Hsuan Boyer, Destiny Computing and Software Systems Division School of Science Technology Engineering and Mathematics University of Washington Bothell BothellWA98011 United States

ISBN: (纸本)9781509061815

Availability of affordable hardware that in effect enables desktop supercomputing has enabled more ambitious neural simulations driven by more complex software. However, this opportunity comes with costs, in terms of long learning curves to take advantage of the performance possibilities of idiosyncratic, architecturally heterogenous hardware and decreasing ability to be confident in the quality of simulation results. This paper describes a new neural simulation and software/data provenance framework that reduces the difficulty of taking full advantage of GPU computing and increases investigator confidence that simulations results are valid. © 2017 IEEE.

关键词： Computer software

来源：评论

学校读者我要写书评

暂无评论

software and data provenance as a basis for eScience workflow 17

Software and data provenance as a basis for eScience workflo...

引用

17th IEEE International Conference on eScience, eScience 2021

作者： Conquest, Joseph Stiber, Michael University of Washington Bothell Computing and Software Systems Division School of Science Technology Engineering and Mathematics BothellWA98011 United States

ISBN: (纸本)9781665403610

While tracking data and software provenance is necessary in eScience, and often implemented in scientific workflow management tools, such tools generally don't provide graphical UIs that use provenance to guide workflow. Graphitti Workbench was built to center workflow around its PROV-DM based ProVis software and data provenance visualizer, making provenance more readily accessible to scientists. Using ProVis as the system desktop highlights the importance of provenance to inform future experiments and improves provenance understandability. Our usability study indicates that this solution is highly acceptable and that it provides easy deployment of experiments using provenance-based creation. © 2021 IEEE.

关键词： e-Science

来源：评论

学校读者我要写书评

暂无评论

DeepTracer-Denoising: Deep Learning for 3D Electron Density Map Denoising

DeepTracer-Denoising: Deep Learning for 3D Electron Density ...

引用

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Guan, Haowen Si, Dong New York University Center of Data Science New York City United States University of Washington Bothell Division of Computing and Software Systems Bothell United States

ISBN: (纸本)9781665468190

Cryo-electron microscopy (Cryo-EM) is widely used in molecular structure determination and drug discovery. Experimental cryo-EM images suffer from the noises introduced by electron beam dose and sample preparation. Although many approaches have been proposed to improve the signal-to-noise ratio (SNR) for cryo-EM image denoising, the noises are still presented after 3D reconstruction and can obstruct the analysis and visualization of the 3D density map. Here we present DeepTracer-Denoising, a method for 3D electron density map denoising. We employ a 3D Neural Network to learn the pattern of noises and the biological structure from density maps. Our method is designed to work on medium to high-resolution maps ranging from 2.5 A to 10.0A. It is configurated with two modes to tackle both background noise and structural noise in a 3D density map. Our method can correctly identify 97.70% background noise while preserving 96.46% density of the native structure. For the maps that contain structural noise, DeepTracer-Denoising achieves an overall accuracy of 98.95%. © 2022 IEEE.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

iCare - An AI - Powered Virtual Assistant for Mental Health

iCare - An AI - Powered Virtual Assistant for Mental Health

引用

IEEE International Conference on Healthcare Informatics (ICHI)

作者： Remya Mavila Sugam Jaiswal Raghav Naswa Weichao Yuwen Bill Erdly Dong Si Computing and Software Systems Division University of Washington Bothell USA

ISBN: (数字)9798350383737

ISBN: (纸本)9798350383744

Currently, mental health issues are increasing among individuals. Sharing feelings with someone who cares plays a key role in resolving these issues. Virtual assistants that can simulate human conversations using artificial intelligence can be effectively used to communicate with individuals facing challenges. This paper describes iCare, a conversational AI that uses machine learning to supply support to individuals suffering from mental health issues. iCare supplies a safe, private, virtual environment for users to share their feelings, and get empathetic response that improves their mental health. It combines multiple techniques to provide the best responses to the users. iCare is designed to provide support for a range of users including those who are suffering from anxiety or depression, and individuals who are unhappy and need help to improve their present feeling.

关键词： Technological innovation Virtual assistants Anxiety disorders Virtual environments Medical treatment Mental health Oral communication

来源：评论

学校读者我要写书评

暂无评论

Identification of Essential Genes with NemoProfile and Various Machine Learning Models 16th

Identification of Essential Genes with NemoProfile and Vario...

引用

16th International Symposium on Bioinformatics Research and Applications, ISBRA 2020

作者： Wang, Yangxiao Kim, Wooyoung Division of Computing and Software Systems School of Science Technology Engineering and Mathematics University of Washington Bothell 18115 Campus Way NE BothellWA98011-8246 United States

ISBN: (纸本)9783030578206

Genes are sequences of nucleotide in DNA that encode proteins. Essential genes are a type of genes that are critical and indispensable for an organism’s survival. Many network-based algorithms have been developed to identify essential genes. We introduce a novel approach to predict essential genes that are based on network motif profiles (NemoProfile) and various machine learning models. Experimental results show that NemoProfile is an effective data feature generated from biological networks, and balanced data is a critical factor to improve the overall performance. © 2020, Springer Nature Switzerland AG.

关键词： Genes

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：