Article
Computer Science, Artificial Intelligence
Ruoxi Deng, Zhao-Min Chen, Huiling Chen, Jie Hu
Summary: In this paper, a novel deep-learning-based method for refining object contours is proposed. By introducing keypoint-focal loss and regularization loss, as well as integrating a Transformer-style hyper module, the proposed method achieves outstanding performance in contour detection tasks.
Article
Chemistry, Multidisciplinary
Dimitrios Zaikis, Christina Karalka, Ioannis Vlahavas
Summary: This article presents a novel approach based on knowledge graph and deep learning techniques for extracting drug-drug interactions from biomedical literature. The proposed method utilizes context information and transfer learning to achieve high classification accuracy.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Suhang Gu, Fu-Lai Chung, Shitong Wang
Summary: The novel algorithm MVCPMM aims to achieve more consistent multi-view clustering results with only one random initialization and one parameter. By adding virtual function nodes to pass smooth messages between different views, MVCPMM improves clustering quality and consistency. Experimental results demonstrate the superiority of MVCPMM in terms of clustering performance and consistency across different views.
INFORMATION SCIENCES
(2022)
Article
Microbiology
Einat Nestor, Gal Toledano, Jonathan Friedman
Summary: To understand microbial communities, knowing all pairwise interactions is crucial. However, measuring these interactions is challenging. Using machine learning, we accurately predicted pairwise interactions between culturable bacteria based on their phylogeny, growth capabilities, and interactions with other species.
Article
Multidisciplinary Sciences
Austin A. Varela, Sammy Cheng, John H. Werren
Summary: ACE2 receptor is crucial for SARS-CoV-2 infection. COVID-19 presents diverse pathologies, including micro-thrombosis, cytokine storms, and inflammatory responses. ACE2 interacts with various proteins that may play a role in COVID-19 pathologies.
Review
Chemistry, Multidisciplinary
Mark John Siringan, Abhiram Dawar, Jianyuan Zhang
Summary: The importance of nanoparticles in the pharmaceutical and biomedical fields has increased due to their attractive surface modification, high drug-loading, and improved pharmacokinetics. Fullerenes, a carbon allotrope, have precise molecular structures, potent radical-scavenging activity, photoactivatable reactive-oxygen species generation, and the ability to confine metal atoms and clusters. They have been applied in various biological contexts, including drug delivery, antioxidative, anti-inflammatory, and photodynamic therapy, and magnetic resonance imaging. This review focuses on the interaction of fullerene materials with biological systems and their structural influence on cellular interactions.
MATERIALS CHEMISTRY FRONTIERS
(2023)
Article
Physics, Condensed Matter
Joseph A. M. Paddison
Summary: Magnetic diffuse scattering, observed above a material's magnetic ordering temperature, contains valuable information about the material's magnetic Hamiltonian. However, lack of appropriate software has limited its utilization. This study presents an open-source program, Spinteract, for efficient refinement of magnetic interaction parameters using powder and single-crystal magnetic diffuse scattering data. Examples of refinements to published experimental data sets are given, along with guidelines for data collection and refinement, and discussions on potential developments of the approach.
JOURNAL OF PHYSICS-CONDENSED MATTER
(2023)
Article
Computer Science, Artificial Intelligence
Sarah Adel Bargal, Andrea Zunino, Vitali Petsiuk, Jianming Zhang, Kate Saenko, Vittorio Murino, Stan Sclaroff
Summary: In state-of-the-art deep single-label classification models, top-k accuracy is usually higher than top-1 accuracy, especially in fine-grained datasets. Exploiting the information provided in the top k predicted classes boosts the final prediction of a model. Guided Zoom proposes a novel way of using explainability to improve model performance.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Review
Plant Sciences
Vittorio Venturi, Cristina Bez
Summary: Next-generation sequencing and computational biology have identified various bacterial groups in plant microbiomes and the mechanisms of cell-cell interactions between bacteria. Research on biotic cell-cell interactions among bacteria in plant microbiomes needs to be accelerated to drive progress in fundamental sciences and translational agriculture for sustainable cultivation of economically important crops.
TRENDS IN PLANT SCIENCE
(2021)
Article
Chemistry, Multidisciplinary
Sioned F. Jones, Himanshu Joshi, Stephen J. Terry, Jonathan R. Burns, Aleksei Aksimentiev, Ulrike S. Eggert, Stefan Howorka
Summary: Equipping DNA with hydrophobic anchors enables targeted interaction with lipid bilayers for various applications. Through experiments and molecular dynamics simulations, the study reveals the complex structure and energetics of hydrophobically tagged DNA within lipid membranes. Fundamental insight gained on DNA-bilayer interactions will guide the rational design of membrane-targeting nanostructures.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2021)
Article
Biochemical Research Methods
Hanyu Zhang, Yunxia Wang, Ziqi Pan, Xiuna Sun, Minjie Mou, Bing Zhang, Zhaorong Li, Honglin Li, Feng Zhu
Summary: Recent studies have shown the significant role of non-coding RNA (ncRNA), particularly lncRNA and miRNA interactions, in biological activities. However, existing methods for identifying novel lncRNA-miRNA interactions have room for improvement in their RNA representation and information extraction approaches. This study proposed a novel method called ncRNAInter, which utilized a comprehensive strategy for RNA representation and an optimized deep learning algorithm to predict lncRNA-miRNA interactions. ncRNAInter showed improved performance compared to existing methods and demonstrated its applicability in different species and disease contexts.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Chemistry, Multidisciplinary
Su-Bin Kim, Na-Kyung Yoo, Soo-Jin Choi
Summary: This study evaluated the effects of interactions between ZnO nanoparticles (NPs) and polyphenols (quercetin and rutin) on cytotoxicity, antioxidant activity, ex vivo intestinal absorption, and solubility. The results showed that these interactions increased the cytotoxicity, intestinal absorption, and solubility of ZnO NPs, but had no effect on the antioxidant activity of the polyphenols. The interactions were more pronounced with quercetin than with rutin.
Article
Computer Science, Artificial Intelligence
Xue Li, Yuanzhi Cheng
Summary: The effectiveness of the message aggregator in graph neural networks is not primarily due to the optimization of edge weights, but rather can be achieved through randomized attention. This finding emphasizes the importance of the network topology in achieving superior performance for message passing iterations.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Riccardo Ievoli, Lucio Palazzo, Giancarlo Ragozini
Summary: Traditional match statistics may not fully reflect the actual tactical style of a team, but by analyzing passing networks and structural features, it is possible to better understand a team's passing behavior. This paper shows how information from passing networks can significantly impact match outcomes, and how certain network variables are related to a team's offensive and finalization actions.
KNOWLEDGE-BASED SYSTEMS
(2021)
Review
Pharmacology & Pharmacy
Yiyuan Kang, Jia Liu, Yanping Jiang, Suhan Yin, Zhendong Huang, Yanli Zhang, Junrong Wu, Lili Chen, Longquan Shao
Summary: The interactions between inorganic-based nanomaterials (NMs) and biological membranes are crucial for developing NM-based therapeutics and addressing nanotoxicology issues. Biological membranes act as platforms mediating NM-immune system contacts, and the challenges and potential applications in these areas are discussed in the overview. Understanding these concepts will enhance the design of therapeutic NM for drug delivery systems by leveraging NMs-membrane interactions and functions.
ADVANCED DRUG DELIVERY REVIEWS
(2021)
Article
Biochemistry & Molecular Biology
Marouen Ben Guebila, Camila M. Lopes-Ramos, Deborah Weighill, Abhijeet Rajendra Sonawane, Rebekka Burkholz, Behrouz Shamsaei, John Platig, Kimberly Glass, Marieke L. Kuijjer, John Quackenbush
Summary: Gene regulation networks play a crucial role in tissue identity, disease development, and therapeutic response. The GRAND database provides computationally-inferred gene regulatory network models and targeting scores for predicting drug effects on network structures and matching potential therapeutic drugs to disease states.
NUCLEIC ACIDS RESEARCH
(2022)
Article
Biochemical Research Methods
Alexander G. McFarland, Nolan W. Kennedy, Carolyn E. Mills, Danielle Tullman-Ercek, Curtis Huttenhower, Erica M. Hartmann
Summary: Identifying variant forms of gene clusters can help infer their evolutionary histories and functions. GeneGrouper is an effective tool for high precision clustering of gene clusters. It can also be used to discover novel gene variations.
Article
Biochemistry & Molecular Biology
Sun-Yang Park, Chitong Rao, Katharine Z. Coyte, Gavin A. Kuziel, Yancong Zhang, Wentao Huang, Eric A. Franzosa, Jing-Ke Weng, Curtis Huttenhower, Seth Rakoff-Nahoum
Summary: This study reveals that the fitness of Bacteroidales in the human gut is determined by a combination of glycans and the metabolite butyrate, and different sugars can activate specific inhibitory functions of butyrate. The fitness effects of butyrate within Bacteroides are mediated by species-level variation in enzyme activity and nucleotide polymorphisms.
Article
Biochemistry & Molecular Biology
Deborah Weighill, Marouen Ben Guebila, Kimberly Glass, John Quackenbush, John Platig
Summary: Understanding how individual genotypes influence gene regulation can improve our understanding of human health and development. EGRET is a method that infers genotype-specific gene regulatory networks to reveal the genetic associations driving complex phenotypes.
Review
Genetics & Heredity
Hannah VanEvery, Eric A. Franzosa, Long H. Nguyen, Curtis Huttenhower
Summary: Studies of the human microbiome have similarities with genome-wide association studies and genetic epidemiology, but also have distinct features and mechanisms that relate to health outcomes. Recent advances have allowed for better understanding of the connections between microbial communities and human health.
NATURE REVIEWS GENETICS
(2023)
Review
Microbiology
Aaron M. Walsh, John Leech, Curtis Huttenhower, Hue Delhomme-Nguyen, Fiona Crispie, Christian Chervaux, Paul D. Cotter
Summary: Molecular technologies, such as high-throughput sequencing, have greatly expanded our understanding of the microbial world, particularly in the context of food fermentation and its impact on human health. This review discusses the history of fermented foods, the advancements in molecular approaches for studying them, and the influence of fermented foods on the gut microbiome. It also explores the potential of bioinformatics to enhance our understanding of these foods.
FEMS MICROBIOLOGY REVIEWS
(2023)
Article
Biochemistry & Molecular Biology
Katherine H. Shutta, Deborah Weighill, Rebekka Burkholz, Marouen Ben Guebila, Dawn L. DeMeo, Helena U. Zacharias, John Quackenbush, Michael Altenbuchinger
Summary: In this study, we propose a network approach based on Gaussian Graphical Models (GGMs) called DRAGON, which allows for the joint analysis of multi-omic data. DRAGON adapts to the differences between omics layers, improving model inference and edge recovery, and can identify key molecular mechanisms.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Multidisciplinary Sciences
Javiera P. Oyarzun, Thomas M. Kuntz, Yoann Stussi, Olivia T. Karaman, Sophia Vranos, Bridget L. Callaghan, Curtis Huttenhower, Joseph E. LeDoux, Elizabeth A. Phelps
Summary: The gut microbiota is associated with individual variability in threat learning, but not extinction learning.
Article
Biochemical Research Methods
Sheila M. Gaynor, Maud Fagny, Xihong Lin, John Platig, John Quackenbush
Summary: This study constructed twenty-nine tissue-specific eQTL networks using GTEx data and evaluated different network specifications. The researchers found that using a thresholded Benjamini-Hochberg q value weighted by the Z-statistic balanced metric reproducibility and computational efficiency. The eQTL networks were complementary to gene regulatory networks in understanding regulation, and highly connected nodes were enriched for tissue-relevant traits.
CELL REPORTS METHODS
(2022)
Article
Health Care Sciences & Services
Matthew Quinn, Arlene Chung, Kimberly Glass
Summary: This article introduces a method called ASCEPT for selecting the optimal set of changepoints in mHealth data. The method involves two stages: identifying statistically significant changepoints and trimming changepoints within linear and seasonal trends. The results show that ASCEPT outperforms other methods and has a significant impact on adjusting changepoints in downstream analysis.
Article
Biotechnology & Applied Microbiology
Siyuan Ma, Dmitry Shungin, Himel Mallick, Melanie Schirmer, Long H. Nguyen, Raivo Kolde, Eric Franzosa, Hera Vlamakis, Ramnik Xavier, Curtis Huttenhower
Summary: Microbiome studies have developed a method for normalization, statistical meta-analysis, and population structure discovery. By applying this method to IBD cohorts, consistent associations and novel taxa have been identified. Additionally, a framework for summarizing population structure has been proposed.
Meeting Abstract
Gastroenterology & Hepatology
Shiying Zhang, Xochitl Morgan, Belgin Dogan, Francois-Pierre J. Martin, Suzy Strickler, Akihiko Oka, Jeremy Herzog, Bo Liu, Scot E. Dowd, Curtis Huttenhower, Matthieu Pichaud, Esra Dogan, Randy Longman, Rhonda Yantiss, Lukas A. Mueller, Ellen J. Scherl, R. Balfour Sartor, Kenneth W. Simpson
Article
Genetics & Heredity
Marouen Ben Guebila, Daniel C. Morgan, Kimberly Glass, Marieke L. Kuijjer, Dawn L. DeMeo, John Quackenbush
Summary: Gene regulatory network inference allows for modeling genome-scale regulatory processes. Researchers have developed a collection of tools to model various regulatory processes and improve their performance through GPU-accelerated calculations.
NAR GENOMICS AND BIOINFORMATICS
(2022)