Article
Geochemistry & Geophysics
Lihui Chen, Zhibing Lai, Gemine Vivone, Gwanggil Jeon, Jocelyn Chanussot, Xiaomin Yang
Summary: To process multispectral (MS) images with arbitrary numbers of bands, we propose a bidirectional recurrent pansharpening network (ArbRPN) that can dynamically reconstruct high-resolution (HR) MS images by adaptively changing the number of recurrence to match the number of bands of the low-resolution (LR) MS images. Additionally, we introduce a mask-based training method (mask-training) to achieve superior performance and robustness during pansharpening. Experimental results demonstrate that our ArbRPN outperforms state-of-the-art (SOTA) methods on MS images with different numbers of bands.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Information Systems
Aytug Onan
Summary: This paper proposes a bidirectional convolutional recurrent neural network architecture for sentiment analysis, which utilizes bidirectional LSTM and GRU layers to extract past and future contexts, and employs a group-wise enhancement mechanism to strengthen important features and weaken less important ones. Experimental results demonstrate that this architecture achieves state-of-the-art performance in sentiment analysis.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Medicine, Research & Experimental
Guiyang Zhang, Qiang Tang, Pengmian Feng, Wei Chen
Summary: In this study, an attention-based bidirectional gated recurrent unit network called IPs-GRUAtt was proposed to identify phosphorylation sites in SARS-CoV-2-infected host cells. Comparative results showed that IPs-GRUAtt outperformed state-of-the-art machine-learning methods and existing models in identifying phosphorylation sites. Moreover, the attention mechanism allowed IPs-GRUAtt to extract key features from protein sequences. These results demonstrate that IPs-GRUAtt is a powerful tool for identifying phosphorylation sites.
MOLECULAR THERAPY-NUCLEIC ACIDS
(2023)
Article
Engineering, Petroleum
Xuechen Li, Xinfang Ma, Fengchao Xiao, Cong Xiao, Fei Wang, Shicheng Zhang
Summary: Machine learning is effective for predicting fractured well production compared to conventional methods, but predicting multistep production remains challenging. To address this, we propose a framework based on bidirectional gated recurrent units and multitask learning, which improves prediction performance by sharing task-dependent representations among multiphase production prediction tasks.
Article
Computer Science, Information Systems
Jose F. Rodrigues-, Marco A. Gutierrez, Gabriel Spadon, Bruno Brandoli, Sihem Amer-Yahia
Summary: This study introduces an artificial neural network architecture, LIG-Doctor, based on two Minimal Gated Recurrent Unit networks, which achieved consistent improvements in prognosis prediction for patients. The results could inspire future research on similar problems.
INFORMATION SCIENCES
(2021)
Article
Engineering, Manufacturing
Yonghwi Kwon, Youngsoo Shin
Summary: The article introduces the use of Recurrent Neural Network (RNN) as a machine learning model for fast optical proximity correction (OPC) and demonstrates the superior accuracy and efficiency of this method through experiments.
IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING
(2021)
Article
Computer Science, Artificial Intelligence
J. Ashok Kumar, S. Abirami, Tina Esther Trueman, Erik Cambria
Summary: Toxicity identification is a serious issue in online communities, and an automatic system like MCBiGRU is proposed for detecting toxic comments. Experimental results show that the MCBiGRU model outperforms in terms of multilabel metrics.
Article
Computer Science, Artificial Intelligence
R. Dharaniya, J. Indumathi, G. V. Uma
Summary: The objective of this study is to perform text generation specifically for movie scripts, identifying context and building scripts through sentiment classification and text vectorization. Bidirectional long short-term memory and multi-head attention mechanism are used to understand future context.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Jieting Chen, Chao Qian, Jie Zhang, Yuetian Jia, Hongsheng Chen
Summary: The authors propose a generation-elimination framework that accurately forecasts inaccessible spectra by correlating spectra from different frequency bands without consulting structural information. This framework accelerates the unification of metasurface designs and enables versatile applications involving cross-wavelength information correlation. The study also introduces a dimensionality reduction approach to visualize the abstract correlated spectra data encoded in latent spaces.
NATURE COMMUNICATIONS
(2023)
Article
Multidisciplinary Sciences
Zhanjie Jing, Xiaohong Gao
Summary: This paper proposes a tailings pond monitoring and early warning system, which utilizes a deep learning network to construct an infiltration line prediction model, aiming to improve the stability and safety management level of tailings ponds.
Article
Computer Science, Information Systems
Abdullah Ali Salamai, Ather Abdulrahman Ageeli, El-Sayed M. El-kenawy
Summary: E-commerce is a system that allows individuals to purchase and sell goods online, aiming to provide convenience to customers by eliminating the need to visit physical stores. This research aims to develop machine learning algorithms for predicting e-commerce sales and tests the proposed algorithm on a time series dataset.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Physics, Multidisciplinary
Manuel S. Rudolph, Ntwali Bashige Toussaint, Amara Katabarwa, Sonika Johri, Borja Peropadre, Alejandro Perdomo-Ortiz
Summary: Generating high-quality data using quantum computers in unsupervised machine learning is an exciting and challenging field. This study presents the first practical implementation of a quantum-classical generative algorithm, which utilizes gate-based quantum computers to generate high-resolution images of handwritten digits. The results show that the quantum-assisted algorithm outperforms comparable classical algorithms in terms of performance.
Article
Engineering, Electrical & Electronic
Guocai Nan, Zhengkuan Wang, Chenghua Wang, Bi Wu, Zhican Wang, Weiqiang Liu, Fabrizio Lombardi
Summary: This work introduces a hybrid-iterative compression algorithm for LSTM/GRU and proposes an energy-efficient accelerator for bidirectional RNNs. By grouping gating units and using different compression algorithms, significant reduction in storage and computation requirements can be achieved without compromising accuracy. Improvements in the data flow of matrix operation unit and BRAM utilization, along with a timing matching strategy, address the load-imbalance issue and result in enhanced energy efficiency compared to state-of-the-art designs.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2021)
Article
Engineering, Electrical & Electronic
Gang Wang, Yanmei Li, Yifei Wang, Zhangjun Wu, Mingfeng Lu
Summary: This article proposes a bidirectional shrinkage gated recurrent unit network with a multiscale attention mechanism for multisensor fault diagnosis. The network combines GRU and soft thresholding denoising strategy to filter out noise-related feature information and employs a multiscale feature learning strategy to learn discriminative multiscale features from nonstationary signals. Experimental results demonstrate that the proposed approach achieves superior diagnostic accuracies and excellent antinoise performance.
IEEE SENSORS JOURNAL
(2023)
Article
Chemistry, Multidisciplinary
Youzhuang Sun, Junhua Zhang, Zhengjun Yu, Yongan Zhang, Zhen Liu
Summary: As a key bridge between logging and seismic data, acoustic(AC) logging data is significant for reservoir lithology, physical property analysis, and quantitative evaluation. Completing AC logging data is challenging due to instrument failure and borehole collapse, but it can be accomplished using other logging parameters.
Article
Automation & Control Systems
Cecile Valsecchi, Magda Collarile, Francesca Grisoni, Roberto Todeschini, Davide Ballabio, Viviana Consonni
Summary: The interest in multitask and deep learning strategies for quantitative structure-activity relationship (QSAR) analysis has been increasing. In this study, the binary classification capability of multitask deep and shallow neural networks were compared to single-task strategies and other benchmark methods. The results showed that multitask learning is beneficial for tasks that are less represented, and multitask deep learning strategies performed similarly to some single-task approaches.
JOURNAL OF CHEMOMETRICS
(2022)
Review
Pharmacology & Pharmacy
Jose Jimenez-Luna, Francesca Grisoni, Nils Weskamp, Gisbert Schneider
Summary: This article reviews the current status of AI in chemoinformatics, discussing topics such as quantitative structure-activity/property relationship and structure-based modeling, de novo molecular design, and chemical synthesis prediction. The advantages and limitations of current deep learning applications are highlighted, offering a perspective on next-generation AI for drug discovery.
EXPERT OPINION ON DRUG DISCOVERY
(2021)
Article
Chemistry, Multidisciplinary
Michael Moret, Moritz Helmstaedter, Francesca Grisoni, Gisbert Schneider, Daniel Merk
Summary: Chemical language models coupled with the beam search algorithm were used to automate molecule design and scoring, resulting in the discovery of novel inverse agonists for retinoic acid receptor-related orphan receptors (RORs). These designs were synthesizable in three reaction steps and exhibited low-micromolar to nanomolar potency towards RORg, showcasing the potential of generative artificial intelligence in data-driven drug discovery.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2021)
Article
Multidisciplinary Sciences
Francesca Grisoni, Berend J. H. Huisman, Alexander L. Button, Michael Moret, Kenneth Atz, Daniel Merk, Gisbert Schneider
Summary: Automating the molecular design-make-test-analyze cycle has led to successful generation of potent LXR agonists, confirming the applicability of the proposed framework for automated drug design.
Article
Chemistry, Medicinal
Michael Moret, Francesca Grisoni, Paul Katzberger, Gisbert Schneider
Summary: Chemical language models (CLMs) are useful for designing molecules with desired properties. This study introduces the perplexity metric to evaluate the generated molecules' similarity to the design objectives, ranking the promising designs. The perplexity scoring also helps identify and remove undesired biases in the model training process.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Medicinal
Derek van Tilborg, Alisa Alenicheva, Francesca Grisoni
Summary: Machine learning plays a crucial role in drug discovery and chemistry. However, the effect of activity cliffs - molecules that are structurally similar but exhibit significant differences in potency - on model performance has received limited attention. In this study, we benchmarked 24 machine and deep learning approaches and found that machine learning methods based on molecular descriptors outperformed more complex deep learning methods in predicting the properties of activity cliffs. Our findings highlight the need for dedicated metrics and novel algorithms to address the limitation posed by activity cliffs in molecular machine learning models.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Multidisciplinary Sciences
Michael Moret, Irene Pachon Angona, Leandro Cotos, Shen Yan, Kenneth Atz, Cyrill Brunner, Martin Baumgartner, Francesca Grisoni, Gisbert Schneider
Summary: Generative chemical language models (CLMs) can be used to generate new molecular structures from a textual representation. Hybrid CLMs can leverage bioactivity information for training compounds. In this study, a virtual compound library was created using a generative CLM and refined using a CLM-based classifier for bioactivity prediction. A new PI3K gamma ligand with sub-micromolar activity was identified, highlighting the potential of hybrid CLMs for molecular design.
NATURE COMMUNICATIONS
(2023)
Review
Biochemistry & Molecular Biology
R. Ozcelik, D. van Tilborg, J. Jimenez-Luna, F. Grisoni
Summary: Artificial intelligence (AI) in the form of deep learning is promising for drug discovery and chemical biology, especially in protein structure prediction, organic synthesis planning, and molecule design. While most efforts have focused on ligand-based approaches, structure-based drug discovery has the potential to address unsolved challenges such as affinity prediction for new protein targets and understanding chemical kinetic properties. Advances in deep learning methodologies and accurate protein structure predictions support a resurgence in structure-based approaches guided by AI. This review summarizes key algorithmic concepts in structure-based deep learning for drug discovery and discusses future opportunities, applications, and challenges.
Article
Biochemistry & Molecular Biology
Francesca Grisoni
Summary: Generative deep learning is revolutionizing de novo drug design by enabling the generation of molecules with specific properties. Chemical language models, which use deep learning to generate new molecules as strings, have been remarkably successful in this endeavor. With advances in natural language processing and interdisciplinary collaborations, chemical language models are expected to play a key role in the future of drug discovery.
CURRENT OPINION IN STRUCTURAL BIOLOGY
(2023)
Correction
Chemistry, Medicinal
Derek van Tilborg, Alisa Alenicheva, Francesca Grisoni
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Medicinal
Marco Ballarotto, Sabine Willems, Tanja Stiller, Felix Nawa, Julian A. A. Marschner, Francesca Grisoni, Daniel Merk
Summary: Generative neural networks trained on SMILES can design innovative bioactive molecules de novo. These models have usually been fine-tuned on template molecules but it is challenging to apply them to orphan targets with few known ligands.
JOURNAL OF MEDICINAL CHEMISTRY
(2023)
Review
Biotechnology & Applied Microbiology
Michael W. Mullowney, Katherine R. Duncan, Somayah S. Elsayed, Neha Garg, Justin J. J. van der Hooft, Nathaniel I. Martin, David Meijer, Barbara R. Terlouw, Friederike Biermann, Kai Blin, Janani Durairaj, Marina Gorostiola Gonzalez, Eric J. N. Helfrich, Florian Huber, Stefan Leopold-Messer, Kohulan Rajan, Tristan de Rond, Jeffrey A. van Santen, Maria Sorokina, Marcy J. Balunas, Mehdi A. Beniddir, Doris A. van Bergeijk, Laura M. Carroll, Chase M. Clark, Djork-Arne Clevert, Chris A. Dejong, Chao Du, Scarlet Ferrinho, Francesca Grisoni, Albert Hofstetter, Willem Jespers, Olga V. Kalinina, Satria A. Kautsar, Hyunwoo Kim, Tiago F. Leao, Joleen Masschelein, Evan R. Rees, Raphael Reher, Daniel Reker, Philippe Schwaller, Marwin Segler, Michael A. Skinnider, Allison S. Walker, Egon L. Willighagen, Barbara Zdrazil, Nadine Ziemert, Rebecca J. M. Goss, Pierre Guyomard, Andrea Volkamer, William H. Gerwick, Hyun Uk Kim, Rolf Mueller, Gilles P. van Wezel, Gerard J. P. van Westen, Anna K. H. Hirsch, Roger G. Linington, Serina L. Robinson, Marnix H. Medema
Summary: The developments in computational omics technologies in combination with artificial intelligence approaches have opened up new possibilities for drug discovery. However, addressing key challenges such as high-quality datasets and algorithm validation is essential to realize the potential of these synergies.
NATURE REVIEWS DRUG DISCOVERY
(2023)
Article
Chemistry, Multidisciplinary
Ana Ortiz-Perez, Cristina Izquierdo-Lozano, Rens Meijers, Francesca Grisoni, Lorenzo Albertazzi
Summary: Barcoding is a powerful tool to distinguish multiple targets within a complex mixture and increase assay throughput. While fluorescent barcoding of microparticles is widely used, it is more challenging for nanoparticles due to their small size and heterogeneity. In this study, a machine-learning-assisted workflow was developed to write, read, and classify barcoded PLGA-PEG nanoparticles at a single-particle level.
NANOSCALE ADVANCES
(2023)
Article
Environmental Sciences
Kamel Mansouri, Agnes L. Karmaus, Jeremy Fitzpatrick, Grace Patlewicz, Prachi Pradeep, Domenico Alberga, Nathalie Alepee, Timothy E. H. Allen, Dave Allen, Vinicius M. Alves, Carolina H. Andrade, Tyler R. Auernhammer, Davide Ballabio, Shannon Bell, Emilio Benfenati, Sudin Bhattacharya, Joyce Bastos, Stephen Boyd, J. B. Brown, Stephen J. Capuzzi, Yaroslav Chushak, Heather Ciallella, Alex M. Clark, Viviana Consonni, Pankaj R. Daga, Sean Ekins, Sherif Farag, Maxim Fedorov, Denis Fourches, Domenico Gadaleta, Feng Gao, Jeffery M. Gearhart, Garett Goh, Jonathan M. Goodman, Francesca Grisoni, Christopher M. Grulke, Thomas Hartung, Matthew Hirn, Pavel Karpov, Alexandru Korotcov, Giovanna J. Lavado, Michael Lawless, Xinhao Li, Thomas Luechtefeld, Filippo Lunghini, Giuseppe F. Mangiatordi, Gilles Marcou, Dan Marsh, Todd Martin, Andrea Mauri, Eugene N. Muratov, Glenn J. Myatt, Dac-Trung Nguyen, Orazio Nicolotti, Reine Note, Paritosh Pande, Amanda K. Parks, Tyler Peryea, Ahsan H. Polash, Robert Rallo, Alessandra Roncaglioni, Craig Rowlands, Patricia Ruiz, Daniel P. Russo, Ahmed Sayed, Risa Sayre, Timothy Sheils, Charles Siegel, Arthur C. Silva, Anton Simeonov, Sergey Sosnin, Noel Southall, Judy Strickland, Yun Tang, Brian Teppen, Igor Tetko, Dennis Thomas, Valery Tkachenko, Roberto Todeschini, Cosimo Toma, Ignacio Tripodi, Daniela Trisciuzzi, Alexander Tropsha, Alexandre Varnek, Kristijan Vukovic, Zhongyu Wang, Liguo Wang, Katrina M. Waters, Andrew J. Wedlake, Sanjeeva J. Wijeyesakere, Dan Wilson, Zijun Xiao, Hongbin Yang, Gergely Zahoranszky-Kohalmi, Alexey Zakharov, Fagen F. Zhang, Zhen Zhang, Tongan Zhao, Hao Zhu, Kimberley M. Zorn, Warren Casey, Nicole C. Kleinstreuer
Summary: The international collaboration in developing in silico models for predicting acute oral toxicity, resulting in the CATMoS, has demonstrated high performance in terms of accuracy and robustness. This modeling suite is being evaluated by regulatory agencies as a potential replacement for in vivo rat acute oral toxicity studies.
ENVIRONMENTAL HEALTH PERSPECTIVES
(2021)
Article
Chemistry, Multidisciplinary
Michael Moret, Moritz Helmstaedter, Francesca Grisoni, Gisbert Schneider, Daniel Merk
Summary: Chemical language models combined with the beam search algorithm as an automated molecule design and scoring technique can generate novel compounds with potential bioactivity. The newly discovered inverse agonists can be synthesized in a few simple reaction steps and exhibit low micromolar to nanomolar potency towards RORg. This model-intrinsic sampling technique eliminates the strict need for external compound scoring functions, further expanding the applicability of generative artificial intelligence to data-driven drug discovery.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2021)