Article
Automation & Control Systems
Dario Passos, Puneet Mishra
Summary: This tutorial aims to provide practical tools and methods for non-expert users in the chemometrics field to learn and implement deep spectral modelling and DL optimization. It showcases two practical examples on implementing and optimizing DL models for spectral regression and classification tasks.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2022)
Article
Biochemical Research Methods
Ya-Ru Miao, Mengxuan Xia, Mei Luo, Tao Luo, Mei Yang, An-Yuan Guo
Summary: ImmuCellAI-mouse is an accurate tool for predicting mouse immune cell abundance, utilizing gene expression data and a hierarchical strategy to accurately predict most immune cell types, revealing dynamic changes in immune cell infiltration in a mouse tumor dataset.
Article
Biochemical Research Methods
Manish Goel, Korbinian Schneeberger
Summary: Third-generation genome sequencing technologies have significantly increased the number of high-quality genome assemblies. Plotsr is an efficient tool for visualizing structural similarities and rearrangements between genomes, capable of comparing on chromosome level or selected regions, and enhancing visualization with regional identifiers or histogram tracks.
Article
Economics
Tasso Adamopoulos, Loren Brandt, Jessica Leight, Diego Restuccia
Summary: Based on household-level panel data from China and a quantitative framework, this study examines the extent and consequences of factor misallocation in agriculture. It finds that within-village frictions in land and capital markets, linked to rural land institutions, disproportionately constrain more productive farmers. These frictions reduce aggregate agricultural productivity by affecting resource allocation across farmers and worker allocation across sectors, particularly in terms of the type of farmers operating in agriculture.
Article
Economics
Daniel Borup, Erik Christian Montes Schutte
Summary: This study demonstrates a strong correlation between Google search activity and employment growth in the United States, showing that Google search results are an effective predictor for future employment growth. By constructing a large panel of 172 variables and utilizing Google's own algorithms to find semantically related search queries, the Google Trends model outperforms other predictors in its ability to forecast employment growth.
JOURNAL OF BUSINESS & ECONOMIC STATISTICS
(2022)
Article
Biochemical Research Methods
Chang Sun, Ping Xuan, Tiangang Zhang, Yilin Ye
Summary: This study proposes a graph convolutional autoencoder and generative adversarial network (GAN) based method (GANDTI) for predicting novel drug-target interactions (DTIs). By constructing a drug-target heterogeneous network and regularizing the feature vectors of nodes into a Gaussian distribution, GANDTI outperforms other methods for drug repositioning.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Sebastian Bittrich, Yana Rose, Joan Segura, Robert Lowe, John D. Westbrook, Jose M. Duarte, Stephen K. Burley
Summary: Membrane proteins, encoded by a significant portion of human genes, account for a majority of FDA-approved drug targets. The RCSB Protein Data Bank web portal has made recent improvements by integrating a wealth of new membrane protein annotations from external resources. These enhancements greatly enhance the presentation of membrane protein data and provide users with tools for searching and visualizing these proteins.
Article
Statistics & Probability
Trevor Hastie, Andrea Montanari, Saharon Rosset, Ryan J. Tibshirani
Summary: This paper studies minimum l(2) norm interpolation least squares regression in the high-dimensional regime, focusing on linear and nonlinear models. The study discovers the phenomena of double descent behavior in prediction risk and potential benefits of overparametrization.
ANNALS OF STATISTICS
(2022)
Article
Biochemical Research Methods
Lihong Peng, Chang Wang, Xiongfei Tian, Liqian Zhou, Keqin Li
Summary: In this study, a deep learning framework with dual-net neural architecture (LPI-DLDN) is developed to identify potential interactions between long non-coding RNAs (lncRNAs) and proteins. Compared with other state-of-the-art prediction methods, LPI-DLDN demonstrates powerful classification performance and reveals new interaction relationships.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Xiaoke Ma, Penggang Sun, Maoguo Gong
Summary: This study proposes an integrative framework for dynamic module detection based on regularized nonnegative matrix factorization method, which can be applied to cancer diagnosis and therapy. Experimental results demonstrate that the framework outperforms other methods in terms of accuracy. For breast cancer data, the obtained dynamic modules are more enriched and can be used to predict cancer stages and patients' survival time.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Kai Cao, Yiguang Hong, Lin Wan
Summary: Motivated by the need for effective approaches to integrate single-cell multi-omics data, this study presents Pamona, a partial Gromov-Wasserstein distance-based manifold alignment framework. It aims to delineate and represent the shared and dataset-specific cellular structures across modalities. Pamona demonstrates superior performance in accurately identifying shared and dataset-specific cells, recovering and aligning cellular structures, outperforming existing methods. The framework also allows for the incorporation of prior information to enhance alignment quality.
Article
Biochemical Research Methods
Lesong Wei, Xiucai Ye, Tetsuya Sakurai, Zengchao Mu, Leyi Wei
Summary: In this study, a novel deep learning framework called ToxIBTL is proposed for predicting the toxicity of peptides and proteins. By utilizing the information bottleneck principle and transfer learning, ToxIBTL effectively retains relevant information and minimizes redundant information in the features, resulting in improved prediction performance.
Article
Biochemical Research Methods
Cheng Yan, Guihua Duan, Yayan Zhang, Fang-Xiang Wu, Yi Pan, Jianxin Wang
Summary: A drug-drug interaction (DDI) refers to the association between drugs where one drug's pharmacological effects are influenced by another drug. This study proposes a novel method, called DDI-IS-SL, to predict DDIs using integrated similarity and semi-supervised learning. DDI-IS-SL combines drug chemical, biological, and phenotype data to calculate the feature similarity of drugs. It also uses a semi-supervised learning method to calculate the interaction possibility scores of drug-drug pairs. DDI-IS-SL demonstrates better prediction performance and shorter computation time compared to other methods, and its performance is further supported by case studies.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Dongyuan Li, Shuyao Zhang, Xiaoke Ma
Summary: Tracking dynamic modules during cancer progression is crucial for studying cancer pathogenesis, diagnosis, and therapy. Current algorithms focusing on detecting dynamic modules without integrating heterogeneous genomic data have limitations, which this study addresses by proposing a novel algorithm (TANMF) that integrates temporal networks and gene attributes. Experimental results demonstrate the superiority of TANMF in accuracy and its ability to identify dynamic modules associated with patients' survival time in breast cancer data.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Review
Engineering, Environmental
Chen Yong, Jinlong Dong, Guo Fei, Tong Bin, Zhou Tao, Fang Hao, Wang Li, Zhan Qinghua
Summary: This study comprehensively examined the research status of landslide susceptibility through visual analysis and summarization. The findings identified the shortcomings in various aspects and proposed future research directions. The results provide valuable references and guidance for understanding the current landslide susceptibility and future studies.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2022)
Article
Statistics & Probability
Ivan Diaz, Nicholas Williams, Katherine L. Hoffman, Edward J. Schenck
Summary: Most causal inference methods focus on fixed values of exposure under interventions, which may not be practical for continuous or multi-valued treatments. Longitudinal modified treatment policies (LMTPs) provide an alternative that yields immediate and practically relevant effects, with an interpretation in terms of meaningful interventions such as changing the exposure by a specific amount. LMTPs also have the advantage of satisfying the positivity assumption required for causal inference.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Statistics & Probability
Jun Yu, HaiYing Wang, Mingyao Ai, Huiming Zhang
Summary: This article proposes a nonuniform subsampling method based on Poisson subsampling to address the computational burden issue for massive data. The optimal subsampling probabilities are derived under the quasi-likelihood estimation and their consistency and asymptotic normality are established. A distributed subsampling framework is also developed for handling data stored in different locations. The effectiveness of the proposed methods is demonstrated through numerical experiments on simulated and real datasets.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Biochemical Research Methods
Cheng-Hong Yang, Kuo-Chuan Wu, Li-Yeh Chuang, Hsueh-Wei Chang
Summary: DNA barcodes are short sequence fragments used for species identification. This study proposes a deep learning framework, called deep barcoding, for species classification using DNA barcodes. By utilizing raw sequence data and deep convolutional neural networks, the deep barcoding model achieves high accuracy in species identification. Although there are challenges, the deep barcoding model has the potential to be an effective tool for species classification.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Engineering, Environmental
Ali O. Alnahit, Ashok K. Mishra, Abdul A. Khan
Summary: This study compares two machine learning methods for predicting water quality parameters in 97 watersheds in the Southeast Atlantic region of the USA. The results show that one algorithm is easier to train and more robust. Furthermore, partial dependence plots highlight the complex and non-linear relationships between predictors and water quality indicators.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2022)
Review
Statistics & Probability
Raphael Huser, Jennifer L. Wadsworth
Summary: The classical modeling of spatial extremes is often too rigid and fails to capture the localized nature of severe events. It is also computationally challenging or less efficient to fit classical spatial extremes models in high dimensions. Recent progress includes new models with flexible tail structures and likelihood-based inference methods for large datasets.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL STATISTICS
(2022)
