Review
Biodiversity Conservation
Freddy A. Diaz-Gonzalez, Jose Vuelvas, Carlos A. Correa, Victoria E. Vallejo, D. Patino
Summary: The use of artificial intelligence techniques to process multidimensional data in agro-industrial systems can help estimate soil quality and provide information on soil management and crop conditions. However, there is currently no identified model for applying these new technologies in medium and low-scale agricultural systems.
ECOLOGICAL INDICATORS
(2022)
Review
Plant Sciences
Junjie Ma, Bangyou Zheng, Yong He
Summary: Recent research in wheat has focused on increasing grain yields and improving grain quality, particularly in terms of protein content and composition. Nitrogen levels during the growing season have a significant impact on these quality factors, and hyperspectral remote sensing is being recognized as an economical alternative for assessing wheat's nitrogen status. Current methods for monitoring nitrogen levels in wheat include hyperspectral vegetation indices and linear nonparametric regression. Machine learning is also being used to model the nonlinear relationship between spectral data and nitrogen status. This study provides a comprehensive review of hyperspectral vegetation indices related to nitrogen and discusses challenges and future directions in evaluating wheat's nitrogen status. The authors also suggest further investigation into the mechanism of protein formation in wheat grains using hyperspectral imaging systems. This overview supports the application of hyperspectral imaging systems in assessing wheat's nitrogen status and has implications for food and nutrition security.
FRONTIERS IN PLANT SCIENCE
(2022)
Article
Meteorology & Atmospheric Sciences
Daniel Ayers, Jack Lau, Javier Amezcua, Alberto Carrassi, Varun Ojha
Summary: Prediction errors grow faster in some situations than in others in chaotic dynamical systems. Real-time knowledge about error growth can enable strategies to adjust modelling and forecasting infrastructure to increase accuracy and reduce computation time. In this feasibility study, supervised machine learning is used to estimate local Lyapunov exponents in place of the classical method. The machine learning algorithms accurately predict stable and unstable local Lyapunov exponents, but only somewhat accurately predict neutral exponents.
QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY
(2023)
Article
Engineering, Environmental
Sang-Soo Baek, Younghun Choi, Junho Jeon, JongCheol Pyo, Jongkwan Park, Kyung Hwa Cho
Summary: The study developed data-driven models to estimate micropollutant concentrations without relying on the SIL standard, with ResNet 101 and RF exhibiting the best performance. The trained models showed accurate training and validation results, providing a potential rapid and economic alternative for micropollutant measurement.
Article
Environmental Sciences
Farnaz Hosseinpour, Naresh Kumar, Trang Tran, Eladio Knipping
Summary: Background ozone refers to ground-level ozone in the absence of any domestic anthropogenic emissions. Recent adjustments to air quality models using data fusion techniques have shown an increased contribution of background ozone, exceeding national air quality standards in some regions.
ATMOSPHERIC ENVIRONMENT
(2024)
Article
Construction & Building Technology
Joana B. Silva, Manar Amayri, Patrick Reignier, Stephane Ploix, Carlos Santos Silva
Summary: A new supervised learning approach is proposed for classifying different instances related to human behavior in an office context, relying on interactive and cooperative learning based on timed data and user feedback. The method ensures privacy by avoiding cameras and introduces a new label correction method for more subjective yet reliable environment classifications. Analyses of different interaction criteria on sensor/feature selection were conducted, along with applications in energy efficiency and demand management.
ENERGY AND BUILDINGS
(2022)
Article
Construction & Building Technology
R. K. Veiga, A. C. Veloso, A. P. Melo, R. Lamberts
Summary: This study demonstrates the application of machine learning to estimate building energy use intensities of bank branches buildings in Brazil. The Support Vector Machine is identified as the most accurate model for benchmarking purposes. By optimizing sophisticated machine learning techniques, the model accuracy is significantly improved, showing high practical value.
ENERGY AND BUILDINGS
(2021)
Article
Multidisciplinary Sciences
Gerard Torrats-Espinosa
Summary: This study examines the impact of racial residential segregation on the spread of COVID-19 in the United States, finding that counties with higher levels of racial segregation have higher mortality and infection rates, as well as racial mortality gaps.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Materials Science, Multidisciplinary
Teng Long, Yixuan Zhang, Nuno M. Fortunato, Chen Shen, Mian Dai, Hongbin Zhang
Summary: We developed an inverse design framework that enables automated generation of stable multicomponent crystal structures and discovered unreported crystal structures through analysis. This method provides convenience for inverse design of multicomponent materials with possible multi-objective optimization.
Article
Biochemistry & Molecular Biology
Rahul Kaushik, Kam Y. J. Zhang
Summary: The structural information of proteins plays a crucial role in understanding their functions and interactions. The protein structure prediction method ProFitFun-Meta combines various structural information to assess the quality of predicted model structures. It has been validated and benchmarked against current state-of-the-art methods, demonstrating its reliability and efficiency. ProFitFun-Meta has the potential to become an integral component of computational pipelines for protein modeling and design.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2022)
Article
Neurosciences
Davood Karimi, Lana Vasung, Camilo Jaimes, Fedel Machado-Rivas, Simon K. Warfield, Ali Gholipour
Summary: This study proposed a data-driven method to improve the accuracy and robustness of white matter fiber orientation distribution function (fODF) estimation, demonstrating effectiveness through training with simulated and real data. In phantom and real data experiments, the method showed higher accuracy compared to other competing methods, particularly in under sampled diffusion measurements. Additionally, expert ratings indicated significantly better reconstruction of various brain tracts using the proposed method.
Article
Environmental Sciences
Saleem Ibrahim, Martin Landa, Ondrej Pesek, Lukas Brodsky, Lena Halounova
Summary: This study developed a machine learning-based scheme using open data to estimate PM2.5 concentrations in Europe. The results showed that air quality improved during the study period, although the lockdown had varying effects in different regions. This research, which is the first to cover the whole of Europe with high spatial and temporal resolutions using open data, has significant implications for future air quality studies.
Article
Mathematics, Interdisciplinary Applications
Vasilis Krokos, Viet Bui Xuan, Stephane P. A. Bordas, Philippe Young, Pierre Kerfriden
Summary: Multiscale computational modelling presents challenges due to high computational costs, but concurrent multiscale methods offer a solution by using cheaper macroscale surrogates. However, microscale problems still present challenges in terms of implementation and cost.
COMPUTATIONAL MECHANICS
(2022)
Article
Geochemistry & Geophysics
T. M. Hansen, C. C. Finlay
Summary: The solution to a probabilistic inverse problem is the posterior probability distribution, which is often sampled due to the difficulty of obtaining a full analytic expression. Decision-makers are interested in the probability of certain features related to the model parameters, and a neural network is proposed to estimate these posterior statistics directly. The method is illustrated on a nonlinear inversion problem with non-Gaussian prior information, and it provides fast results comparable to slower sampling methods.
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2022)
Article
Geochemistry & Geophysics
T. M. Hansen, C. C. Finlay
Summary: This study demonstrates how to use a neural network to directly estimate posterior statistics of the features of the posterior distribution, overcoming the high computational costs associated with traditional sampling methods.
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2022)
Article
Chemistry, Medicinal
Yoshikazu Nishimura, Toru Esaki, Yoshiaki Isshiki, Naoki Okamoto, Yoshiyuki Furuta, Tomoya Kotake, Yoshiaki Watanabe, Masateru Ohta, Toshito Nakagawa, Hiroshi Noda, Masaru Shimizu, Hitoshi Saito, Tatsuya Tamura, Haruhiko Sato
JOURNAL OF MEDICINAL CHEMISTRY
(2018)
Article
Chemistry, Medicinal
Yoshikazu Nishimura, Toru Esaki, Yoshiaki Isshiki, Yoshiyuki Furuta, Akemi Mizutani, Tomoya Kotake, Takashi Emura, Yoshiaki Watanabe, Masateru Ohta, Toshito Nakagawa, Kotaro Ogawa, Shinichi Arai, Hiroshi Noda, Hidetomo Kitamura, Masaru Shimizu, Tatsuya Tamura, Haruhiko Sato
JOURNAL OF MEDICINAL CHEMISTRY
(2020)
Article
Chemistry, Multidisciplinary
Ryosuke Kojima, Shoichi Ishida, Masateru Ohta, Hiroaki Iwata, Teruki Honma, Yasushi Okuno
JOURNAL OF CHEMINFORMATICS
(2020)
Article
Chemistry, Medicinal
Koichiro Kato, Tomohide Masuda, Chiduru Watanabe, Naoki Miyagawa, Hideo Mizouchi, Shumpei Nagase, Kikuko Kamisaka, Kanji Oshima, Satoshi Ono, Hiroshi Ueda, Atsushi Tokuhisa, Ryo Kanada, Masateru Ohta, Mitsunori Ikeguchi, Yasushi Okuno, Kaori Fukuzawa, Teruki Honma
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2020)
Article
Chemistry, Multidisciplinary
Takashi Yoshidome, Mitsunori Ikeguchi, Masateru Ohta
JOURNAL OF COMPUTATIONAL CHEMISTRY
(2020)
Article
Multidisciplinary Sciences
Shuntaro Chiba, Aki Tanabe, Makoto Nakakido, Yasushi Okuno, Kouhei Tsumoto, Masateru Ohta
SCIENTIFIC REPORTS
(2020)
Article
Chemistry, Medicinal
Kei Takedomi, Masateru Ohta, Toru Ekimoto, Mitsunori Ikeguchi
Summary: The S810L mutation in the mineralocorticoid receptor (MR) causes a change in progesterone's biological activity, leading to disease. Molecular dynamics simulations indicate that the inflow of water molecules may affect the dissociation process of progesterone in different types of MR.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Article
Biochemistry & Molecular Biology
Satomi Kori, Yuki Shibahashi, Toru Ekimoto, Atsuya Nishiyama, Sae Yoshimi, Kosuke Yamaguchi, Satoru Nagatoishi, Masateru Ohta, Kouhei Tsumoto, Makoto Nakanishi, Pierre-Antoine Defossez, Mitsunori Ikeguchi, Kyohei Arita
Summary: Accumulation of epigenetic alterations is a major cause of tumorigenesis, with drugs targeting DNA methylation-regulating factors showing potential for cancer therapy. UHRF1, essential for DNA methylation maintenance, is overexpressed in cancer cells, making it a promising therapeutic target. The discovery of 5A-DMP as a novel TTD-binding compound and inhibitor of full-length UHRF1:LIG1 interaction suggests potential for future cancer therapy experiments.
BIOORGANIC & MEDICINAL CHEMISTRY
(2021)
Article
Chemistry, Medicinal
Kosuke Kawama, Yusaku Fukushima, Mitsunori Ikeguchi, Masateru Ohta, Takashi Yoshidome
Summary: In this study, a deep learning model is proposed to rapidly estimate the distribution functions and position of water molecules around proteins, with a prediction time more than two orders of magnitude faster than traditional methods. The predicted results show good agreement with the actual distribution functions obtained using traditional methods.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Medicinal
Tatsuya Yoshizawa, Shoichi Ishida, Tomohiro Sato, Masateru Ohta, Teruki Honma, Kei Terayama
Summary: Designing highly selective molecules for drug targets is challenging. In this study, a structure generator based on reinforcement learning was developed to optimize multiple objectives and successfully proposed selective inhibitors for tyrosine kinases.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Multidisciplinary
Yugo Shimizu, Masateru Ohta, Shoichi Ishida, Kei Terayama, Masanori Osawa, Teruki Honma, Kazuyoshi Ikeda
Summary: In this study, a new molecule-generation method that takes into account the patent status of molecules was developed. This method enables the efficient generation of novel molecules with high drug-likeness and will help in the effective development of drug compounds.
JOURNAL OF CHEMINFORMATICS
(2023)
Meeting Abstract
Biophysics
Kosuke Kawama, Yusaku Fukushima, Takashi Yoshidome, Mitsunori Ikeguchi, Masateru Ohta
BIOPHYSICAL JOURNAL
(2022)
Correction
Multidisciplinary Sciences
Shuntaro Chiba, Aki Tanabe, Makoto Nakakido, Yasushi Okuno, Kouhei Tsumoto, Masateru Ohta
SCIENTIFIC REPORTS
(2021)
Meeting Abstract
Biophysics
Takashi Yoshidome, Mitsunori Ikeguchi, Masateru Ohta
BIOPHYSICAL JOURNAL
(2021)