Article
Multidisciplinary Sciences
Yang Li, Zheng Wang, Li-Ping Li, Zhu-Hong You, Wen-Zhun Huang, Xin-Ke Zhan, Yan-Bin Wang
Summary: This study introduces a computational method for predicting PPIs based on protein sequence information, utilizing a combination of OLPP and RoF models to identify non-interacting and interacting protein pairs with high accuracy on Yeast and Human datasets. The proposed method serves as a valuable tool in accelerating the resolution of key problems in proteomics.
SCIENTIFIC REPORTS
(2021)
Article
Biochemical Research Methods
Min Li, Zhangli Lu, Yifan Wu, YaoHang Li
Summary: In this study, an end-to-end neural network model called BACPI is proposed to predict compound-protein interactions (CPIs) and binding affinity. The model combines graph attention network and convolutional neural network (CNN) to learn representations of compounds and proteins, and uses a bi-directional attention neural network to integrate these representations. The results show that the BACPI model outperforms other machine learning methods in predicting CPIs and achieves higher performance in predicting binding affinities compared to other state-of-the-art deep learning methods.
Article
Computer Science, Artificial Intelligence
Kahyun Jeon, Ghang Lee, Seoungwoo Kang, Hyunsung Roh, Jeaeun Jung, Kyungha Lee, Mark Baldwin
Summary: This study proposes a relational framework to address the lack of commonly accepted standard data schema in IDM specifications, laying the foundation for developing an international standard for IDM data schema.
ADVANCED ENGINEERING INFORMATICS
(2021)
Article
Business
Cuiqing Jiang, Yiru Zhou, Bo Chen
Summary: Financial distress prediction has traditionally relied on accounting features from financial statements, but this study proposes a framework for extracting statistical and semantic features from patent text. The study finds that patent features contain incremental information related to financial distress, expanding the feature space of financial distress research. This research has implications for loan approval, investment decision-making, and patent pledges.
TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE
(2023)
Article
Multidisciplinary Sciences
Maryam Khalid, Akane Sano
Summary: This study predicts self-reported happiness and stress levels using mobile sensing data, incorporating individual physiology and environmental factors through weather and social networks. A machine learning architecture is developed to aggregate information from the graph network and predict emotion for all users. Constructing social networks does not incur additional costs or raise privacy concerns.
SCIENTIFIC REPORTS
(2023)
Editorial Material
Biochemistry & Molecular Biology
Samuel J. Taylor, Sriram Sundaravel, Ulrich Steidl
Summary: The study reveals that AML cells have a specific addiction to the IRF8MEF2D gene expression network, and identifies ZMYND8 as a chromatin reader that regulates the IRF8-MEF2D program critical for AML cell survival.
Article
Geochemistry & Geophysics
Kuai Dai, Chi Ma, Zhaolin Wang, Yongshen Long, Xutao Li, Shanshan Feng, Yunming Ye
Summary: This study proposes a hierarchical spatial-temporal network (HSTnet) for satellite image sequence prediction. HSTnet can learn effective spatial-temporal features by extracting features at both pixel level and patch level. It also introduces a dual-branch Transformer to capture patch-level spatial and temporal features. Experimental results show that HSTnet outperforms state-of-the-art approaches on the satellite dataset.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Computer Science, Information Systems
Junran Wu, Ke Xu, Xueyuan Chen, Shangzhe Li, Jichang Zhao
Summary: This study proposes a novel framework to address the issues of long-term dependencies and chaotic properties in stock prediction. By transforming time series into complex networks and extracting structural information from the mapped graphs, the performance of the prediction model is improved. The effectiveness of the framework is validated through real-world stock data and trading simulations.
INFORMATION SCIENCES
(2022)
Article
Business
Hannes W. W. Lampe
Summary: This article investigates the impact of crowdsourcing for knowledge creation on decision-making in patent application examination, focusing on the issue of local search bias among patent examiners. The study examines the Peer To Patent initiative of USPTO, which opens up the patent examination process to public participation for the first time. The findings provide empirical evidence that crowdsourcing helps overcome the local search bias of examiners, leading to a greater reliance on atypical and less formalized knowledge.
Article
Chemistry, Medicinal
Xiaowei Shen, Shiding Zhang, Jianyu Long, Changjing Chen, Meng Wang, Ziheng Cui, Biqiang Chen, Tianwei Tan
Summary: Determining the catalytic site is important for understanding the relationship between protein sequence, structure, and function. A novel model using a graph neural network has been developed for predicting enzyme catalytic sites. The model achieved high accuracy and outperformed existing models on benchmark datasets. This research provides a valuable tool for understanding protein sequence-structure-function relationships and characterizing novel enzymes.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Computer Science, Artificial Intelligence
Jingjing Gu, Qiang Zhou, Jingyuan Yang, Yanchi Liu, Fuzhen Zhuang, Yanchao Zhao, Hui Xiong
Summary: This paper proposes an interpretable framework for bike flow prediction (IBFP) in dockless bike sharing systems. By modeling bike flows and extracting traffic patterns, the IBFP method can effectively predict bike flows with interpretability. Experimental results demonstrate the advantages of IBFP in flow prediction, and a case study further illustrates the interpretability of the flow pattern exploitation.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Information Systems
Longjiang Li, Jie Wang, Rui Zhang, Yuanchen Gao, Yonggang Li, Yuming Mao
Summary: This article proposes a highly exposure-resilient framework that protects the privacy of a single key by fusing multiple redundant keys and significantly reduces the secrecy outage probability.
IEEE SYSTEMS JOURNAL
(2022)
Article
Computer Science, Hardware & Architecture
Rui Tang, Xingshu Chen, Chuancheng Wei, Qindong Li, Wenxian Wang, Haizhou Wang, Wei Wang
Summary: This paper proposes an interlayer link prediction framework based on multiple structural attributes (MulAtt) that calculates the matching degree of unmatched nodes once by leveraging the information of closed triad, intralayer links, matched neighbors, and intralayer links of neighbors simultaneously to ensure accuracy while reducing time consumption. The framework achieves better performance than several existing network structure-based methods in a non-iterative way.
Article
Chemistry, Analytical
Shaohua Zhou, Cheng Yang, Jian Wang
Summary: This paper studies the impact of amplifier specification degradation on system failure, and models the temperature characteristics of amplifiers using an extreme learning machine (ELM). The results show that the model agrees well with the measurement results and can effectively reduce measurement time and cost.
Article
Biotechnology & Applied Microbiology
Yongfei Qin, Chao Li, Xia Shi, Weigang Wang
Summary: This study focuses on the estrogen receptor ER alpha, which plays a key role in the development of breast cancer and is considered an important target for treatment. The researchers developed a compound activity prediction model using statistical regression and neural network methods. The model successfully predicted the bioactivity values of new compounds, providing valuable insights for the development of anti-breast cancer drugs.
FRONTIERS IN BIOENGINEERING AND BIOTECHNOLOGY
(2022)
Article
Biochemistry & Molecular Biology
Jenny Sandmark, Anna Tigerstrom, Tomas Akerud, Magnus Althage, Thomas Antonsson, Stefan Blaho, Cristian Bodin, Jonas Bostrom, Yantao Chen, Anders Dahlen, Per-Olof Eriksson, Emma Evertsson, Tomas Fex, Ola Fjellstrom, David Gustafsson, Margareta Herslof, Ryan Hicks, Emelie Jarkvist, Carina Johansson, Inge Kalies, Birgitta Karlsson Svalstedt, Fredrik Kartberg, Anne Legnehed, Sofia Martinsson, Andreas Moberg, Marianne Ridderstrom, Birgitta Rosengren, Alan Sabirsh, Anders Thelin, Johanna Vinblad, Annika U. Wellner, Bingze Xu, Ann-Margret Ostlund-Lindqvist, Wolfgang Knecht
JOURNAL OF BIOLOGICAL CHEMISTRY
(2020)
Article
Chemistry, Multidisciplinary
Josep Arus-Pous, Atanas Patronov, Esben Jannik Bjerrum, Christian Tyrchan, Jean-Louis Reymond, Hongming Chen, Ola Engkvist
JOURNAL OF CHEMINFORMATICS
(2020)
Article
Biochemical Research Methods
Mei Ding, Christian Tyrchan, Elisabeth Back, Jorgen Ostling, Steffen Schubert, Christopher McCrae
Article
Biochemistry & Molecular Biology
Izaskun Mitxitorena, Domenico Somma, Jennifer P. Mitchell, Matti Lepisto, Christian Tyrchan, Emma L. Smith, Patrick A. Kiely, Helen Walden, Karen Keeshan, Ruaidhri J. Carmody
JOURNAL OF BIOLOGICAL CHEMISTRY
(2020)
Review
Pharmacology & Pharmacy
Anna Tomberg, Jonas Bostroem
DRUG DISCOVERY TODAY
(2020)
Article
Chemistry, Multidisciplinary
Hafeez S. Haniff, Laurent Knerr, Xiaohui Liu, Gogce Crynen, Jonas Bostrom, Daniel Abegg, Alexander Adibekian, Elizabeth Lekah, Kye Won Wang, Michael D. Cameron, Ilyas Yildirim, Malin Lemurell, Matthew D. Disney
Article
Chemistry, Medicinal
Fabio Begnini, Vasanthanathan Poongavanam, Bjorn Over, Marie Castaldo, Stefan Geschwindner, Patrik Johansson, Mohit Tyagi, Christian Tyrchan, Lisa Wissler, Peter Sjo, Stefan Schiesser, Jan Kihlberg
Summary: This study demonstrates the use of macrocyclic natural product cores as a high-quality in silico screening library for lead generation for challenging drug targets. By iteratively docking a selected set of natural product-derived cores, an uncharged macrocyclic inhibitor of the Keap1-Nrf2 protein-protein interaction was discovered. This inhibitor shows cellular efficacy and potential for further optimization.
JOURNAL OF MEDICINAL CHEMISTRY
(2021)
Article
Chemistry, Multidisciplinary
Jiazhen He, Huifang You, Emil Sandstrom, Eva Nittinger, Esben Jannik Bjerrum, Christian Tyrchan, Werngard Czechtizky, Ola Engkvist
Summary: The study focuses on molecular optimization in drug discovery using machine translation models to generate molecules with desirable properties. By incorporating the concept of matched molecular pairs and user-specified property changes into the models, molecules satisfying specific requirements can be effectively generated.
JOURNAL OF CHEMINFORMATICS
(2021)
Article
Chemistry, Multidisciplinary
Daniel Fernandez-Llaneza, Silas Ulander, Dea Gogishvili, Eva Nittinger, Hongtao Zhao, Christian Tyrchan
Summary: Activity prediction is crucial in drug discovery, and the SiameseCHEM model demonstrates superior performance in predicting the bioactivity of small molecules compared to traditional machine learning models, especially when handling SMILES strings.
Article
Chemistry, Medicinal
Giuseppina La Sala, Anders Gunnarsson, Karl Edman, Christian Tyrchan, Anders Hogner, Andrey Frolov
Summary: The study uses a combination of methods to investigate the structural and dynamic features of allosteric cross-talk within the glucocorticoid receptor (GR), identifying a network of evolutionarily conserved residues involved in signal transduction. Molecular dynamics simulations clarify the dynamic interconnections within this network, offering a mechanistic explanation for how different peptides affect the intensity of the allosteric signal. This research provides valuable insights into the allosteric regulation of GR and lays a foundation for designing novel drugs.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Article
Chemistry, Medicinal
Igor Shamovsky, Lena Ripa, Frank Narjes, Britta Bonn, Stefan Schiesser, Ina Terstiege, Christian Tyrchan
Summary: This study investigates the mechanism of N-hydroxylation of aromatic and heteroaromatic amines by CYP1A2 using density functional theory calculations, finding that bioactivation follows an anionic pathway. The results demonstrate that mutagenicity of ArNH2 can be removed by disrupting geometric and electrostatic fit to CYP1A2.
JOURNAL OF MEDICINAL CHEMISTRY
(2021)
Article
Chemistry, Medicinal
Dhanushka Weerakoon, Rodrigo J. Carbajo, Leonardo De Maria, Christian Tyrchan, Hongtao Zhao
Summary: In this study, the conformational behavior of two PROTACs, MZ1 and dBET6, was investigated using molecular dynamics simulations. The simulations revealed different linker conformational behaviors and a tendency towards an intramolecular lipophilic interaction between the two warheads. The dissociation mechanism of the von Hippel-Lindau-MZ1-BRD4 complex was also studied and shown to follow a two-step pathway.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Medicinal
Kosala N. Amarasinghe, Leonardo De Maria, Christian Tyrchan, Leif A. Eriksson, Jens Sadowski, Dusan Petrovic
Summary: Peptides have been widely used in drug discovery, but current peptide optimization mainly focuses on a small number of natural and commercially available non-natural amino acids. However, the chemical spaces for small molecule drug discovery are much larger. In this study, researchers developed a large virtual library of readily synthesizable non-natural amino acids to aid in virtual screening and peptide optimization. Through computational simulations, they demonstrated that the non-natural amino acid chemical space can be massively extended and virtually screened with a reasonable computational cost.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Multidisciplinary Sciences
Martina Audagnotto, Werngard Czechtizky, Leonardo De Maria, Helena Kack, Garegin Papoian, Lars Tornberg, Christian Tyrchan, Johan Ulander
Summary: Proteins can exist in different conformations, and recent findings have shown that co-evolutionary analysis combined with machine-learning techniques can improve distance predictions between residue pairs. In this study, deep learning approaches and mechanistic modeling were used to investigate protein conformational ensembles. The predicted models were compared to experimental structures, and a potential correlation between experimental dynamics and predicted models was found, highlighting the areas of improvement in protein folding and dynamics prediction.
SCIENTIFIC REPORTS
(2022)
Article
Chemistry, Multidisciplinary
Karolina Kwapien, Eva Nittinger, Jiazhen He, Christian Margreitter, Alexey Voronov, Christian Tyrchan
Summary: This study examines the predictability of four properties relevant for drug design using different data sets and machine learning algorithms. The study confirms that additive data are the easiest to predict, highlighting the complexity of predicting nonadditive events in drug design.