Article
Engineering, Multidisciplinary
Hao Liu, Dechang Pi, Shuyuan Qiu, Xixuan Wang, Chang Guo
Summary: A data-driven identification model of the associated fault propagation path is proposed in this paper. The model utilizes KPCA method for fault detection, introduces a new transfer entropy method for constructing causality diagram, and utilizes a kernel extreme learning machine-based search approach for fault propagation path identification. Experimental results show that the proposed method obtains a lower number of causal edges compared to traditional and multivariate methods.
Article
Engineering, Mechanical
S. Safari, J. M. Londono Monsalve
Summary: A methodology for structural identification of nonlinear assemblies is proposed, which enables the discovery of stiffness and damping nonlinear models even in unmeasured locations. This methodology takes into account dominant modal couplings and is demonstrated on a case study of a nonlinear structure with a frictional bolted joint, showing successful model selection and parameter estimation for weakly nonlinear elements.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Engineering, Multidisciplinary
Kazem Meidani, Amir Barati Farimani
Summary: This study introduces a machine learning method to discover terms in partial differential equations from spatiotemporal data. By extracting robust physical features from data samples, the method can effectively represent the behaviors imposed by each mathematical term. In comparison to previous models, this approach allows for the discovery of 2D equations with different orders of time derivatives and identification of new underlying physics.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2021)
Article
Engineering, Civil
Huan Luo, Stephanie German Paal
Summary: This article proposes a novel data-driven method to predict lateral seismic response based on estimated lateral stiffness, integrating machine learning approach with the hysteretic model. The method significantly outperforms classical methods in both prediction capabilities and computational efficiency.
EARTHQUAKE SPECTRA
(2022)
Article
Medicine, Research & Experimental
Yan-Chak Li, Hsiao-Hsien Leon Hsu, Yoojin Chun, Po-Hsiang Chiu, Zoe Arditi, Luz Claudio, Gaurav Pandey, Supinda Bunyavanich
Summary: The study utilized machine learning methods to investigate combinations of early-life air toxics associated with asthma, revealing multiple air toxic combinations significantly associated with asthma outcomes.
JOURNAL OF CLINICAL INVESTIGATION
(2021)
Review
Plant Sciences
Rui Zheng, Yang Sun, Xiaoyu Zhang, Chen Zhao, Pengqian Wang, Shiqi Chen, Zhao Chen, Ruijin Qiu, Aihua Liang, Hongcai Shang
Summary: This comprehensive review examines the clinical features of adverse events (AEs) associated with the combination of XYP and RB. The review analyzes data from randomized controlled trials, cohort studies, case-control studies, case reports, case series, and the National Adverse Drug Reaction Monitoring Information System. The most common AE reported is skin and appendage reactions, with a majority of cases being pseudo-allergic reactions. The study recommends increased awareness of the safety of the XYP-RB combination treatment, especially in children, and the standardization of medication protocols.
JOURNAL OF ETHNOPHARMACOLOGY
(2022)
Article
Engineering, Civil
Gao Fan, Jun Li, Hong Hao, Yu Xin
Summary: This paper introduces a Segment based Conditional Generative Adversarial Network (SegGAN) for structural dynamic response reconstruction, showing superior accuracy and noise immunity over DenseNet and traditional CNN through numerical studies.
ENGINEERING STRUCTURES
(2021)
Article
Engineering, Civil
Debarshi Sen, James Long, Hao Sun, Xander Campman, Oral Buyukozturk
Summary: This paper discusses the development of a surrogate model for predicting structural response using a multi-component deconvolution seismic interferometry approach in cases of sparse sensor deployment and limited information about the structure of interest. The proposed algorithm considers various sources of uncertainties and demonstrates its effectiveness by applying it to field monitoring data collected from structures with sparse sensor deployment in the Groningen region of the Netherlands.
ENGINEERING STRUCTURES
(2021)
Article
Biochemistry & Molecular Biology
Mengying Han, Sheng Liu, Dachuan Zhang, Rui Zhang, Dongliang Liu, Huadong Xing, Dandan Sun, Linlin Gong, Pengli Cai, Weizhong Tu, Junni Chen, Qian-Nan Hu
Summary: This study presents the construction of the comprehensive database AddictedChem for rapid identification of new psychoactive substances (NPS). Predictive models were created using machine learning algorithms and molecular descriptors, achieving high accuracy for NPS identification. A consensus strategy was used to identify potential NPS. Additionally, a chemical space was constructed to better understand the existence of NPS.
Article
Agricultural Engineering
Xin Wang, Yu Zhang, Suyuan Jia, Haoyu Deng, Wenbiao Xu, Junyou Shi
Summary: The effective utilization of lignin is crucial for sustainable biomass conversion and the development of renewable resources. This study focuses on the catalytic oxidative transformation of lignin and explores its structural features and their impact on product diversity. The content of beta-O-4 linkages in lignin and the presence of S-type structural units are found to be important factors that influence the yield and efficiency of the conversion process. However, the complex nature of lignin presents challenges for its efficient depolymerization and utilization. A comprehensive understanding of lignin's structure is crucial for advancing its effective utilization and the development of renewable resources.
INDUSTRIAL CROPS AND PRODUCTS
(2024)
Editorial Material
Biotechnology & Applied Microbiology
Andrew S. Robertson, Alexis Reisin Miller, Felipe Dolz
Summary: Drug developers are increasingly using data-driven analysis to understand regulatory agencies' expectations, but are limited by the availability and quality of regulatory datasets. Establishing a single, robust FDA regulatory actions database could help address this limitation.
NATURE REVIEWS DRUG DISCOVERY
(2021)
Article
Obstetrics & Gynecology
Minuo Yin, Jiaming Zhang, Xinliu Zeng, Hanke Zhang, Ying Gao
Summary: This study identified significant gene modules related to OE, involving the PI3K/Akt and aging pathways. ITPR1 expression was found to be elevated in OE lesions and its knockdown inhibited cell proliferation and induced apoptosis in HESCs. Candidate drugs camptothecin and irinotecan were identified as promising for OE treatment, with camptothecin suppressing HESC proliferation and inducing apoptosis. The study also validated the therapeutic effect of camptothecin in an OE mouse model.
FERTILITY AND STERILITY
(2021)
Article
Chemistry, Physical
Weihan Li, Iskender Demir, Decheng Cao, Dominik Joest, Florian Ringbeck, Mark Junker, Dirk Uwe Sauer
Summary: This study develops a data-driven parameter identification framework for electrochemical models of lithium-ion batteries in real-world operations using artificial intelligence. The framework improves the accuracy of parameter identification and overcomes the overfitting problem caused by limited battery data.
ENERGY STORAGE MATERIALS
(2022)
Review
Biochemistry & Molecular Biology
G. Beis, A. P. Serafeim, I. Papasotiriou
Summary: Over the past few decades, drug discovery has improved patient outcomes, but challenges remain for rapid development of novel drugs. This article describes a bioinformatic approach for identifying novel cancer drug targets through statistical analysis and co-expression networks. It also provides an overview of druggability assessment methodologies to prioritize and select the best targets to pursue.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Multidisciplinary Sciences
Yuichi Shiraishi, Ai Okada, Kenichi Chiba, Asuka Kawachi, Ikuko Omori, Raul Nicolas Mateos, Naoko Iida, Hirofumi Yamauchi, Kenjiro Kosaki, Akihide Yoshimi
Summary: This study developed a method to detect genomic variants causing splicing changes using transcriptome data alone. By applying this method to a large dataset, the researchers identified a significant number of genomic variants associated with intron retention. Additionally, by exploring the positional relationships with known disease variants, they extracted potential disease-associated variants.
NATURE COMMUNICATIONS
(2022)
Article
Hematology
Heikki Kuusanmaki, Sari Kytola, Ida Vanttinen, Tanja Ruokoranta, Amanda Ranta, Jani Huuhtanen, Minna Suvela, Alun Parsons, Annasofia Holopainen, Anu Partanen, Milla E. L. Kuusisto, Sirpa Koskela, Riikka Raty, Maija Itala-Remes, Imre Vastrik, Olli Dufva, Sanna Siitonen, Kimmo Porkka, Krister Wennerberg, Caroline A. Heckman, Pia Ettala, Marja Pyorala, Johanna Rimpilainen, Timo Siitonen, Mika Kontro
Summary: The BCL-2 inhibitor venetoclax has revolutionized the treatment of acute myeloid leukemia (AML) in patients not benefiting from intensive chemotherapy. However, treatment failure remains a challenge, and predictive markers are needed, particularly for relapsed or refractory AML. Ex vivo drug sensitivity testing may correlate with outcomes, but its prospective predictive value remains unexplored.
Article
Oncology
Sarang S. Talwelkar, Mikko I. Mayranpaa, Julia Schuler, Nora Linnavirta, Annabrita Hemmes, Simone Adinolfi, Matti Kankainen, Wolfgang Sommergruber, Anna-Liisa Levonen, Jari Rasanen, Aija Knuuttila, Emmy W. Verschuren, Krister Wennerberg
Summary: Treatment with ALK inhibitors improves outcome for NSCLC patients with ALK-rearranged tumors, but resistance typically develops. In this study, tumor cell cultures were generated from an ALK-rearranged tumor specimen and drug screens identified a role for PI3K beta and EGFR inhibition in enhancing ALK-inhibitor response and preventing resistance. Combinatorial treatment with ALK and PI3K beta inhibitors showed promise in targeting ALK-rearranged NSCLC.
MOLECULAR ONCOLOGY
(2023)
Article
Medicine, Research & Experimental
Johanna Huttunen, Thales Kronenberger, Ahmed B. Montaser, Adela Kralova, Tetsuya Terasaki, Antti Poso, Kristiina M. Huttunen
Summary: This study explored the interactions between L-type amino acid transporter 1 (LAT1)-utilizing prodrugs and sodium-coupled neutral amino acid transporter 2 (SNAT2). It was found that the cellular uptake of LAT1-utilizing prodrugs in MCF-7 cells was mediated by SNATs, which increased at higher pH, decreased in the absence of sodium, and was inhibited by a SNAT-inhibitor. Docking, molecular dynamics simulations, and analysis confirmed the chemical features supporting the interactions of the studied compounds with SNAT2.
MOLECULAR PHARMACEUTICS
(2023)
Article
Medicine, Research & Experimental
Katayun Bahrami, Juulia Järvinen, Tuomo Laitinen, Mika Reinisalo, Paavo Honkakoski, Antti Poso, Kristiina M. Huttunen, Jarkko Rautio
Summary: In this study, a series of LAT1-targeted drug-phenylalanine conjugates were evaluated. Through in vitro studies and induced-fit docking, it was concluded that smaller compounds were preferred for uptake by LAT1. The flexibility of the ligand played a crucial role in determining the transportability and interactions with LAT1. Introducing polar groups enhanced interactions, while compounds with a carbamate bond in the para-position of the aromatic ring displayed efficient transport efficiencies. The findings of this study have implications for designing CNS or antineoplastic drug candidates and discovering LAT1 inhibitors for cancer therapy.
MOLECULAR PHARMACEUTICS
(2023)
Article
Computer Science, Theory & Methods
Zhirong Yang, Yuwei Chen, Denis Sedov, Samuel Kaski, Jukka Corander
Summary: Neighbor Embedding (NE) is an effective principle for data visualization, but current methods may hide large-scale patterns. To address this, we propose a new cluster visualization method based on the NE principle and present a family of NE methods that can better display clusters.
STATISTICS AND COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Sebastiaan De Peuter, Antti Oulasvirta, Samuel Kaski
Summary: We need to rethink how AI can assist designers by supporting their creativity and problem-solving capabilities instead of automating their tasks. The challenge is to understand designers' goals and provide help without interrupting their workflow. AI-assisted design introduces a framework based on generative user models to infer and adapt to designers' goals, reasoning, and capabilities.
Article
Chemistry, Medicinal
Teodor Dimitrov, Athina Anastasia Moschopoulou, Lennart Seidel, Thales Kronenberger, Mark Kudolo, Antti Poso, Christian Geibel, Pascal Woelffing, Daniel Dauch, Lars Zender, Dieter Schollmeyer, Juergen Bajorath, Michael Forster, Stefan Laufer
Summary: The ATM kinase is an important regulator of the cellular response to DNA double-strand breaks and is considered a promising target in cancer treatment. This study introduces a new class of specific benzimidazole-based ATM inhibitors with high potency against the isolated enzyme and favorable selectivity within related kinases. These inhibitors show strong enzymatic and cellular activities, as well as promising pharmacokinetic properties and selectivities within the PIKK and PI3K families.
JOURNAL OF MEDICINAL CHEMISTRY
(2023)
Article
Pharmacology & Pharmacy
Glaucio Valdameri, Diogo Henrique Kita, Julia de Paula Dutra, Diego Lima Gomes, Arun Kumar Tonduru, Thales Kronenberger, Bruno Gavinho, Izadora Volpato Rossi, Mariana Mazetto de Carvalho, Basile Peres, Ingrid Fatima Zattoni, Fabiane Gomes de Moraes Rego, Geraldo Picheth, Rilton Alves de Freitas, Antti Poso, Suresh V. Ambudkar, Marcel Ramirez, Ahcene Boumendjel, Vivian Rotuno Moure
Summary: Inhibition of ABC transporters is a promising strategy to overcome multidrug resistance in cancer. This study identified a potent ABCG2 inhibitor called chromone 4a (C4a) that exhibits selectivity towards ABCG2. C4a effectively inhibited the efflux of different substrates mediated by ABCG2 and showed potential for drug delivery using liposomes and extracellular vesicles (EVs) in overcoming poor water solubility.
Article
Chemistry, Medicinal
Glaucio Monteiro Ferreira, Thales Kronenberger, Vinicius Goncalves Maltarollo, Antti Poso, Fernando de Moura Gatti, Vitor Medeiros Almeida, Sandro Roberto Marana, Carla Duque Lopes, Daiane Yukie Tezuka, Sergio de Albuquerque, Flavio da Silva Emery, Gustavo Henrique Goulart Trossini
Summary: The etiological agent of Chagas disease, Trypanosoma cruzi, relies on precise epigenetic regulation during host transitions. In this study, we used molecular modelling and experimental validation to discover new inhibitors from commercially available compound libraries. Six inhibitors were selected from virtual screening and validated on the recombinant Sir2 enzyme. The most potent inhibitor (CDMS-01, IC50 = 40 mu M) was chosen as a potential lead compound.
Article
Computer Science, Artificial Intelligence
Mustafa Mert Celikok, Pierre-Alexandre Murena, Samuel Kaski
Summary: Modeling has aimed to eliminate human involvement for objectivity and automation. However, this has limited modeling workflows to well-defined problems. Reintroducing humans into models through iterative processes and collaborative AI assistance could broaden the scope of modeling. Advancements in scoping modeling problems and user models are necessary to realize this vision of human-centric machine learning pipelines.
FRONTIERS IN ARTIFICIAL INTELLIGENCE
(2023)
Article
Chemistry, Medicinal
Mohammad Moein, Markus Heinonen, Natalie Mesens, Ronnie Chamanza, Chidozie Amuzie, Yvonne Will, Hugo Ceulemans, Samuel Kaski, Dorota Herman
Summary: Drug-induced liver injury (DILI) is a significant safety concern and a major cause for drug market withdrawal. Recent advancements in machine learning methods have allowed the development of various in silico models for DILI prediction based on molecule chemical structures. This study introduces a novel phenotype-based annotation approach, using hepatotoxicity information obtained from repeated dose in vivo preclinical toxicology studies, to create a more informative and reliable dataset for machine learning algorithms. The dataset consists of 430 unique compounds with diverse liver pathology findings, which were utilized to develop multiple DILI prediction models trained on publicly available data (TG-GATES) using compound fingerprints. The study demonstrates accurate prediction of DILI labels of TG-GATES compounds and explores the impact of differences between TG-GATES and external test compounds (Johnson & Johnson) on model generalization performance.
CHEMICAL RESEARCH IN TOXICOLOGY
(2023)
Article
Chemistry, Medicinal
Toni Sivula, Laxman Yetukuri, Tuomo Kalliokoski, Heikki Kasnanen, Antti Poso, Ina Pohner
Summary: The emergence of ultra-large screening libraries poses a challenge for docking-based virtual screening. Machine learning-boosted strategies like HASTEN combine rapid ML prediction with the brute-force docking to increase screening throughput. In our case study, we observed a significant reduction in docking experiments by 99% using HASTEN, resulting in shorter screening time.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Chemistry, Organic
Xiaodan Ouyang, Paul M. D'Agostino, Matti Wahlsten, Endrews Delbaje, Jouni Jokela, Perttu Permi, Greta Gaiani, Antti Poso, Piia Bartos, Tobias A. M. Gulder, Hannu Koistinen, David P. Fewer
Summary: In this study, a comparative bioinformatic analysis was used to identify radiosumin biosynthetic gene clusters in the genomes of 13 filamentous cyanobacteria. The entire biosynthetic gene cluster was captured and expressed in Escherichia coli. High-resolution liquid chromatography-mass spectrometry, nuclear magnetic resonance spectroscopy, and chemical degradation analysis revealed the chemical structure of novel radiosumins produced by cyanobacteria. Radiosumin C was found to inhibit human trypsin isoforms selectively.
ORGANIC & BIOMOLECULAR CHEMISTRY
(2023)
Article
Chemistry, Medicinal
Karoline B. Waitman, Larissa C. de Almeida, Marina C. Primi, Jorge A. E. G. Carlos, Claudia Ruiz, Thales Kronenberger, Stefan Laufer, Marcia Ines Goettert, Antti Poso, Sandra V. Vassiliades, Vinicius A. M. de Souza, Monica F. Z. J. Toledo, Neuza M. A. Hassimotto, Michael D. Cameron, Thomas D. Bannister, Leticia Costa-Lotufo, Joa o A. Machado-Neto, Mauricio T. Tavares, Roberto Parise-Filho
Summary: A series of hybrid inhibitors combining pharmacophores of known kinase inhibitors and benzohydroxamate HDAC inhibitors were synthesized and evaluated for their anticancer activity and pharmacokinetic properties. Compounds 4d-f exhibited promising cytotoxicity against hematological cells and moderate activity against solid tumor models. Compound 4d showed potent inhibition of multiple kinase targets and had stable interactions with HDAC and members of the JAK family. These compounds showed selective cytotoxicity with minimal effects on non-tumorigenic cells and favorable pharmacokinetic profiles.
EUROPEAN JOURNAL OF MEDICINAL CHEMISTRY
(2024)
Meeting Abstract
Biochemistry & Molecular Biology
Sophie Wharrie, Zhiyu Yang, Vishnu Raj, Rahul Gupta, Remo Monti, Ying Wang, Pier Francesco Palamara, Samuel Kaski, Andrea Ganna, Christoph Lippert, Pekka Marttinen
EUROPEAN JOURNAL OF HUMAN GENETICS
(2023)