Patterns最新文献_第3页

A deep learning model for characterizing protein-RNA interactions from sequences at single-base resolution. 从单碱基分辨率的序列中表征蛋白质- rna相互作用的深度学习模型。

IF 6.7

Patterns Pub Date : 2025-01-10 DOI: 10.1016/j.patter.2024.101150

Xilin Shen, Yayan Hou, Xueer Wang, Chunyong Zhang, Jilei Liu, Hongru Shen, Wei Wang, Yichen Yang, Meng Yang, Yang Li, Jin Zhang, Yan Sun, Kexin Chen, Lei Shi, Xiangchun Li

{"title":"A deep learning model for characterizing protein-RNA interactions from sequences at single-base resolution.","authors":"Xilin Shen, Yayan Hou, Xueer Wang, Chunyong Zhang, Jilei Liu, Hongru Shen, Wei Wang, Yichen Yang, Meng Yang, Yang Li, Jin Zhang, Yan Sun, Kexin Chen, Lei Shi, Xiangchun Li","doi":"10.1016/j.patter.2024.101150","DOIUrl":"10.1016/j.patter.2024.101150","url":null,"abstract":"Protein-RNA interactions play pivotal roles in regulating transcription, translation, and RNA metabolism. Characterizing these interactions offers key insights into RNA dysregulation mechanisms. Here, we introduce Reformer, a deep learning model that predicts protein-RNA binding affinity from sequence data. Trained on 225 enhanced cross-linking and immunoprecipitation sequencing (eCLIP-seq) datasets encompassing 155 RNA-binding proteins across three cell lines, Reformer achieves high accuracy in predicting binding affinity at single-base resolution. The model uncovers binding motifs that are often undetectable through traditional eCLIP-seq methods. Notably, the motifs learned by Reformer are shown to correlate with RNA processing functions. Validation via electrophoretic mobility shift assays confirms the model's precision in quantifying the impact of mutations on RNA regulation. In summary, Reformer improves the resolution of RNA-protein interaction predictions and aids in prioritizing mutations that influence RNA regulation.","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"6 1","pages":"101150"},"PeriodicalIF":6.7,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11783876/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143081132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Benchmarking the robustness of the correct identification of flexible 3D objects using common machine learning models. 使用通用机器学习模型对正确识别柔性3D物体的鲁棒性进行基准测试。

IF 6.7

Patterns Pub Date : 2025-01-10 DOI: 10.1016/j.patter.2024.101147

Yang Zhang, Andreas Vitalis

{"title":"Benchmarking the robustness of the correct identification of flexible 3D objects using common machine learning models.","authors":"Yang Zhang, Andreas Vitalis","doi":"10.1016/j.patter.2024.101147","DOIUrl":"10.1016/j.patter.2024.101147","url":null,"abstract":"True three-dimensional (3D) data are prevalent in domains such as molecular science or computer vision. In these data, machine learning models are often asked to identify objects subject to intrinsic flexibility. Our study introduces two datasets from molecular science to assess the classification robustness of common model/feature combinations. Molecules are flexible, and shapes alone offer intra-class heterogeneities that yield a high risk for confusions. By blocking training and test sets to reduce overlap, we establish a baseline requiring the trained models to abstract from shape. As training data coverage grows, all tested architectures perform better on unseen data with reduced overfitting. Empirically, 2D embeddings of voxelized data produced the best-performing models. Evidently, both featurization and task-appropriate model design are of continued importance, the latter point reinforced by comparisons to recent, more specialized models. Finally, we show that the shape abstraction learned from database samples extends to samples that are evolving explicitly in time.","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"6 1","pages":"101147"},"PeriodicalIF":6.7,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11783895/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143081150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Why the Environmental Data & Governance Initiative is archiving public environmental data. 为什么“环境数据与治理倡议”要归档公共环境数据？

IF 6.7

Patterns Pub Date : 2025-01-10 DOI: 10.1016/j.patter.2024.101151

Eric Nost, Gretchen Gehrke, Lourdes Vera, Steve Hansen

引用次数: 0

PhosNetVis: A web-based tool for fast kinase-substrate enrichment analysis and interactive 2D/3D network visualizations of phosphoproteomics data. PhosNetVis：一个基于网络的工具，用于快速激酶-底物富集分析和磷酸化蛋白质组学数据的交互式2D/3D网络可视化。

IF 6.7

Patterns Pub Date : 2025-01-10 DOI: 10.1016/j.patter.2024.101148

Osho Rawal, Berk Turhan, Irene Font Peradejordi, Shreya Chandrasekar, Selim Kalayci, Sacha Gnjatic, Jeffrey Johnson, Mehdi Bouhaddou, Zeynep H Gümüş

{"title":"PhosNetVis: A web-based tool for fast kinase-substrate enrichment analysis and interactive 2D/3D network visualizations of phosphoproteomics data.","authors":"Osho Rawal, Berk Turhan, Irene Font Peradejordi, Shreya Chandrasekar, Selim Kalayci, Sacha Gnjatic, Jeffrey Johnson, Mehdi Bouhaddou, Zeynep H Gümüş","doi":"10.1016/j.patter.2024.101148","DOIUrl":"10.1016/j.patter.2024.101148","url":null,"abstract":"Protein phosphorylation involves the reversible modification of a protein (substrate) residue by another protein (kinase). Liquid chromatography-mass spectrometry studies are rapidly generating massive protein phosphorylation datasets across multiple conditions. Researchers then must infer kinases responsible for changes in phosphosites of each substrate. However, tools that infer kinase-substrate interactions (KSIs) are not optimized to interactively explore the resulting large and complex networks, significant phosphosites, and states. There is thus an unmet need for a tool that facilitates user-friendly analysis, interactive exploration, visualization, and communication of phosphoproteomics datasets. We present PhosNetVis, a web-based tool for researchers of all computational skill levels to easily infer, generate, and interactively explore KSI networks in 2D or 3D by streamlining phosphoproteomics data analysis steps within a single tool. PhostNetVis lowers barriers for researchers by rapidly generating high-quality visualizations to gain biological insights from their phosphoproteomics datasets. It is available at https://gumuslab.github.io/PhosNetVis/.","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"6 1","pages":"101148"},"PeriodicalIF":6.7,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11783894/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143081293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How we measure our impact. 我们如何衡量我们的影响。

IF 6.7

Patterns Pub Date : 2025-01-10 DOI: 10.1016/j.patter.2024.101152

Andrew L Hufton

引用次数: 0

Neuron signal attenuation activation mechanism for deep learning. 深度学习神经元信号衰减激活机制。

IF 6.7

Patterns Pub Date : 2024-12-16 eCollection Date: 2025-01-10 DOI: 10.1016/j.patter.2024.101117

Wentao Jiang, Heng Yuan, Wanjun Liu

引用次数: 0

Integration of large language models and federated learning. 大型语言模型和联邦学习的集成。

IF 6.7

Patterns Pub Date : 2024-12-13 DOI: 10.1016/j.patter.2024.101098

Chaochao Chen, Xiaohua Feng, Yuyuan Li, Lingjuan Lyu, Jun Zhou, Xiaolin Zheng, Jianwei Yin

{"title":"Integration of large language models and federated learning.","authors":"Chaochao Chen, Xiaohua Feng, Yuyuan Li, Lingjuan Lyu, Jun Zhou, Xiaolin Zheng, Jianwei Yin","doi":"10.1016/j.patter.2024.101098","DOIUrl":"https://doi.org/10.1016/j.patter.2024.101098","url":null,"abstract":"As the parameter size of large language models (LLMs) continues to expand, there is an urgent need to address the scarcity of high-quality data. In response, existing research has attempted to make a breakthrough by incorporating federated learning (FL) into LLMs. Conversely, considering the outstanding performance of LLMs in task generalization, researchers have also tried applying LLMs within FL to tackle challenges in relevant domains. The complementarity between LLMs and FL has already ignited widespread research interest. In this review, we aim to deeply explore the integration of LLMs and FL. We propose a research framework dividing the fusion of LLMs and FL into three parts: the combination of LLM sub-technologies with FL, the integration of FL sub-technologies with LLMs, and the overall merger of LLMs and FL. We first provide a comprehensive review of the current state of research in the domain of LLMs combined with FL, including their typical applications, integration advantages, challenges faced, and future directions for resolution. Subsequently, we discuss the practical applications of the combination of LLMs and FL in critical scenarios such as healthcare, finance, and education and provide new perspectives and insights into future research directions for LLMs and FL.","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"5 12","pages":"101098"},"PeriodicalIF":6.7,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11701858/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142956218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data-knowledge co-driven innovations in engineering and management. 数据知识共同驱动的工程和管理创新。

IF 6.7

Patterns Pub Date : 2024-12-13 DOI: 10.1016/j.patter.2024.101114

Yingji Xia, Xiqun Michael Chen, Sudan Sun

引用次数: 0

Decorrelative network architecture for robust electrocardiogram classification. 鲁棒心电图分类的去相关网络结构。

IF 6.7

Patterns Pub Date : 2024-12-09 eCollection Date: 2024-12-13 DOI: 10.1016/j.patter.2024.101116

Christopher Wiedeman, Ge Wang

引用次数: 0

Best holdout assessment is sufficient for cancer transcriptomic model selection. 最佳抵抗评估是充分的癌症转录组模型选择。

IF 6.7

Patterns Pub Date : 2024-12-06 eCollection Date: 2024-12-13 DOI: 10.1016/j.patter.2024.101115

Jake Crawford, Maria Chikina, Casey S Greene

{"title":"Best holdout assessment is sufficient for cancer transcriptomic model selection.","authors":"Jake Crawford, Maria Chikina, Casey S Greene","doi":"10.1016/j.patter.2024.101115","DOIUrl":"https://doi.org/10.1016/j.patter.2024.101115","url":null,"abstract":"Guidelines in statistical modeling for genomics hold that simpler models have advantages over more complex ones. Potential advantages include cost, interpretability, and improved generalization across datasets or biological contexts. We directly tested the assumption that small gene signatures generalize better by examining the generalization of mutation status prediction models across datasets (from cell lines to human tumors and vice versa) and biological contexts (holding out entire cancer types from pan-cancer data). We compared model selection between solely cross-validation performance and combining cross-validation performance with regularization strength. We did not observe that more regularized signatures generalized better. This result held across both generalization problems and for both linear models (LASSO logistic regression) and non-linear ones (neural networks). When the goal of an analysis is to produce generalizable predictive models, we recommend choosing the ones that perform best on held-out data or in cross-validation instead of those that are smaller or more regularized.","PeriodicalId":36242,"journal":{"name":"Patterns","volume":"5 12","pages":"101115"},"PeriodicalIF":6.7,"publicationDate":"2024-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11701843/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142956206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0