Article
Chemistry, Medicinal
Philippe Gantzer, Benoit Creton, Carlos Nieto-Draghi
Summary: This paper addresses the problem of molecular structure generation by revisiting fragment-based methods and genetic-based methods, defining a set of indices to compare generation methods. These indices provide information on the coverage of data space, the representativeness of data space exploration, and the generation specificity of molecules without the use of a database composed of real chemicals as a reference.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2021)
Article
Biochemical Research Methods
Ricardo M. M. Borges, Gabriela de Assis Ferreira, Mariana Martins Campos, Andrew Magno Teixeira, Fernanda das Neves Costa, Fernanda Oliveira Chagas
Summary: We developed a tool called DBsimilarity that organizes structure databases into similarity networks, making it easier for natural product chemists to visualize information. The tool converts SDF files into CSV files, calculates compound similarities, and constructs similarity networks in CSV format. Using DBsimilarity, several potential antibiotic compounds were identified in Ginkgo biloba compounds, along with other compounds worthy of further investigation.
PHYTOCHEMICAL ANALYSIS
(2023)
Article
Chemistry, Multidisciplinary
Hiroyuki Kuwahara, Xin Gao
Summary: The study identified 2D fingerprints with little to no contribution to the eigenvalue distribution of the feature matrix using an eigenvalue-based entropy approach, showing that their presence can substantially affect molecular similarity scores and bias the outcome of molecular similarity analysis. This has implications for the optimal selection of 2D fingerprints for compound similarity analysis and identification of potential hits for compounds with target biological activity in virtual screening.
JOURNAL OF CHEMINFORMATICS
(2021)
Article
Mathematics
Alina Barbulescu, Lucica Barbes, Cristian Stefan Dumitriu
Summary: This study aims to analyze the similarity degree of molecules belonging to two subgroups of Aminoalkylindoles through Cheminformatics methods. The Tanimoto coefficients were computed to reveal the level of similarity, and dendrograms and heatmaps were built. The study also found atom-pair similarities within the same group and validated the clustering using the Kruskal-Wallis test.
Article
Chemistry, Medicinal
Onur Caki, Bilge Karacali
Summary: In this study, a Quasi-Supervised Learning (QSL) algorithm is proposed to address the issue of predicting compound-protein interactions. Experimental results show that the proposed method can effectively identify actual interactions.
MOLECULAR INFORMATICS
(2022)
Review
Biochemical Research Methods
Zhongqi Wen, Cheng Yan, Guihua Duan, Suning Li, Fang-Xiang Wu, Jianxin Wang
Summary: This study provides a comprehensive overview of existing methods for predicting human microbe-disease associations, including data sources, method classification, and performance evaluation. Based on computational principles and experimental results, the advantages and disadvantages of these methods are discussed, and suggestions for improvement are proposed.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Chemistry, Multidisciplinary
Sara Szymkuc, Tomasz Badowski, Bartosz A. Grzybowski
Summary: Organic chemistry has seen exponential growth in terms of molecules and specific reactions, but the discovery of new reaction classes/types is slower, indicating a growing reliance on known methods. The newly discovered chemistries are more complex, enabling rapid construction of complex structures in fewer steps.
ANGEWANDTE CHEMIE-INTERNATIONAL EDITION
(2021)
Article
Chemistry, Medicinal
Ramon Alain Miranda-Quintana, David Bajusz, Anita Racz, Karoly Heberger
Summary: The study investigates the conditions under which two comparative measures provide equivalent results on a given set of molecules. A novel method (Differential Consistency Analysis) is introduced to study the consistency between comparative measures, revealing that using a reference with less variation in similarity or representing molecules in a size-independent way can improve consistency. The presented derivations are applicable to all binary similarity coefficients introduced so far, regardless of molecular representations.
MOLECULAR INFORMATICS
(2021)
Article
Chemistry, Physical
Igor Baskin, Yair Ein-Eli
Summary: Electrochemoinformatics, as a scientific discipline, applies information technologies, such as data science, machine learning, and artificial intelligence, to solve problems in electrochemical processes. It is closely related to chemoinformatics and materials informatics and primarily used in battery science and technology.
ADVANCED ENERGY MATERIALS
(2022)
Article
Plant Sciences
Luis Aurelio Montoya-Inzunza, Aldo Moreno-Ulloa, Rommel A. Carballo-Castaneda, Jorge Xool-Tamayo, Laura Aracely Contreras-Angulo, Nayely Leyva-Lopez, Marilena Antunes-Ricardo, Jose Reyes Gonzalez-Galaviz, Jose Basilio Heredia, Erick Paul Gutierrez-Grijalva
Summary: This study characterized the metabolome of Phoradendron brachystachyum plant and identified specific metabolites responsible for lipid- and glucose-lowering effects. Leaf extracts showed the highest activity and were rich in phenolic compounds, flavonoids, and tannins.
Review
Polymer Science
Ioan Szava, Sorin Vlase, Maria Luminita Scutaru, Zsolt Asztalos, Botond-Pal Galfi, Adrian Soica, Simona Soica
Summary: This review discusses the utilization of dimensional methods to address problems in additive manufacturing processes. It points out that this field has not fully explored the potential of these methods. The focus is mainly on polymer-based AM technology, exploring the design and optimization of highly competitive products and layer deposition.
Article
Chemistry, Multidisciplinary
Masoud Reyhani Hamedani, Sang-Wook Kim
Summary: This study investigates the effectiveness and efficiency of graph embedding methods in the task of node similarity computation by comparing them with similarity measures. The results show that, in most datasets, embedding methods are less effective and efficient than similarity measures, and require more time-consuming parameter tuning due to having more parameters. Additionally, increasing the number of dimensions does not necessarily improve the effectiveness of embedding methods in computing node similarity.
APPLIED SCIENCES-BASEL
(2021)
Article
Biochemistry & Molecular Biology
Leonardo Bernal, Luca Pinzi, Giulio Rastelli
Summary: Prostate cancer (PC) is a common cancer in males with poor prognosis in advanced stages. Current therapeutic options are limited and not very effective. This study analyzed compounds with anti-proliferative activity on PC cell lines and identified potential drugs for repurposing against PC.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Biochemical Research Methods
Tingting Zhao, Guangyu Zhu, Harsh Vardhan Dubey, Patrick Flaherty
Summary: Large-scale multiple perturbation experiments can provide a detailed understanding of molecular pathways in response to genetic and environmental changes. A method based on the model-X knockoffs framework and Deep Neural Networks is proposed to identify significant gene expression changes. Identification of genes that respond to specific perturbation stressors can improve the understanding of disease mechanisms and aid in the identification of new drug targets.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Tingting Jiang, Chenyang Bu, Yi Zhu, Xindong Wu
Summary: The objective of entity alignment is to determine if entities refer to the same object in the real world. This paper introduces two groups of methods, symbol-based and embedding-based, for entity alignment. It proposes a promising strategy by combining the advantages of both methods and presents an improved algorithm, ESEA. Experimental results demonstrate that ESEA outperforms other embedding-based methods and the previous RTEA method.
PATTERN RECOGNITION
(2022)