Article
Biochemical Research Methods
Peng Zhou, Li Wen, Jing Lin, Li Mei, Qian Liu, Shuyong Shang, Juelin Li, Jianping Shu
Summary: This study develops a general-purpose method for modeling and predicting the binding affinities of protein-peptide interactions. The method combines unsupervised and supervised approaches to create an integrated PpI affinity predictor. The predictor performs well in calculating affinities from complex structures but can only qualitatively predict globally docked structures.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Chemistry, Multidisciplinary
Katsuhisa Matsumoto, Tomoyuki Miyao, Kimito Funatsu
Summary: In this study, a ranking-oriented QSAR model was utilized for activity prediction in ligand-based drug design, showing promising results when compared to traditional methods. Accumulated experimental data and rigorous validation demonstrated that models trained on compounds from similar assays were equally effective as those trained on compounds from all assays.
Article
Agriculture, Multidisciplinary
Nancy D. Asen, Chibuike C. Udenigwe, Rotimi E. Aluko
Summary: The aim of this study was to determine the structural requirements for peptides that inhibit acetylcholinesterase and butyrylcholinesterase activities. A dataset of 19 oligopeptides identified through mass spectrometry was used to analyze the structure-function relationship. The study found that the most active peptides had specific amino acid combinations at specific positions.
JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY
(2023)
Article
Biochemical Research Methods
Tri Minh Nguyen, Thin Nguyen, Truyen Tran
Summary: By incorporating interaction information from related tasks, the proposed method shows advantages in predicting drug-target interactions compared to other pre-training methods.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemistry & Molecular Biology
Jeff Gaither, Yi-Hsuan Lin, Ralf Bundschuh
Summary: The study focuses on the interactions of hundreds of RNA binding proteins in the human genome with RNA in cells, introducing RBPBind as a web-based tool for quantitatively predicting the interaction by considering the effect of RNA secondary structure on binding affinity. The tool provides a quick and easy way to obtain reliable predicted binding affinities and locations for single-stranded RNA binding proteins based solely on RNA sequence.
JOURNAL OF MOLECULAR BIOLOGY
(2022)
Article
Chemistry, Multidisciplinary
Stefan M. Kohlbacher, Thierry Langer, Thomas Seidel
Summary: QSAR methods are commonly used in drug discovery, where pharmacophores provide advantageous properties for building quantitative SAR models that can generalize to different datasets with low requirements.
JOURNAL OF CHEMINFORMATICS
(2021)
Article
Chemistry, Multidisciplinary
Maogang Li, Weipeng Lai, Ruirui Li, Jiajun Zhou, Yingzhe Liu, Tao Yu, Tianlong Zhang, Hongsheng Tang, Hua Li
Summary: With the development of green chemistry, the new generation of energetic materials exhibit higher insensitivity, density and energy. An ensemble modeling strategy combining Monte Carlo (MC) and variable importance measurement (VIM) improved random forest (RF) and quantitative structure-property relationship (QSPR) was proposed for accurate prediction of detonation performance of energetic materials, which was successfully applied to density prediction.
Article
Engineering, Chemical
Gergo Ignacz, Gyorgy Szekely
Summary: Methods for determining solute rejection in organic solvent nanofiltration are time-consuming and expensive. This study presents two prediction methods based on quantitative structure-activity relationship using traditional machine learning and deep learning models, providing a new platform for a more in-depth investigation into membrane-solute interactions.
JOURNAL OF MEMBRANE SCIENCE
(2022)
Article
Biochemistry & Molecular Biology
Jurgen Drewe, Ernst Kuesters, Felix Hammann, Matthias Kreuter, Philipp Boss, Verena Schoening
Summary: The AMPK plays a critical role in regulating important cellular functions and activation of AMPK is beneficial in various diseases. Machine learning algorithms can accurately predict AMPK activators, making them useful tools for rapid screening and identification of potential compounds.
Article
Pharmacology & Pharmacy
Domenico Gadaleta, Luca D'Alessandro, Marco Marzo, Emilio Benfenati, Alessandra Roncaglioni
Summary: The thyroid system plays a crucial role in physiological processes but can be disrupted by xenobiotics and contaminants, leading to various diseases. This study introduces QSAR models to predict the TPO inhibitory potential, developed using machine learning methods and validated rigorously internally and externally.
FRONTIERS IN PHARMACOLOGY
(2021)
Review
Food Science & Technology
Zhen Ma, Anqi Guo, Pu Jing
Summary: This article reviews the protein-anthocyanin interactions and their impact on the functional and nutritional values of anthocyanin-containing foods. The current understanding of protein-anthocyanin interactions is summarized, along with an outline of the binding characterization of dietary protein-anthocyanin complexes. Furthermore, the article critically discusses the advances in understanding the structure-affinity relationship of protein-anthocyanin interactions and evaluates the associated properties of these complexes for their functional and nutritional values.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
(2022)
Article
Biochemistry & Molecular Biology
Xu Hong, Xiaoxue Tong, Juan Xie, Pinyu Liu, Xudong Liu, Qi Song, Sen Liu, Shiyong Liu
Summary: Understanding protein-RNA interaction is crucial for structural biology. The binding or dissociation process between protein and RNA dominates the regulatory networks in organisms. Developing computational methods to predict the binding affinity for protein-RNA complexes is necessary due to the time-consuming and labor-intensive nature of experimental determination.
PROTEINS-STRUCTURE FUNCTION AND BIOINFORMATICS
(2023)
Article
Biochemical Research Methods
Jialin Wu, Zhe Liu, Xiaofeng Yang, Zhanglin Lin
Summary: In this study, a deep learning model based on a multi-objective neural network was proposed for predicting compound-protein interaction site and binding affinity. The results demonstrated improvements in prediction and better performance in binding site prediction.
BMC BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Hanxiao Xu, Da Xu, Naiqian Zhang, Yusen Zhang, Rui Gao
Summary: This study proposed a PPI prediction algorithm using amino acid sequence information, general regression neural network, and two feature extraction methods, achieving high prediction accuracy on different datasets. Additionally, experimental results showed that the performance of the new method was significantly better than previous methods.
JOURNAL OF PROTEOME RESEARCH
(2021)
Article
Biochemistry & Molecular Biology
Yan Liu, Fan Zhang, Ling Jiang, J. Jefferson P. Perry, Zhihe Zhao, Jiayu Liao
Summary: This study introduces a novel method to determine kinetics parameters of product inhibition using quantitative FRET(qFRET) assay, providing a convenient way to determine all kinetics parameters and a new approach to combine different measurements with mutually compatible results and errors for the first time.
INTERNATIONAL JOURNAL OF BIOLOGICAL MACROMOLECULES
(2021)
Review
Biochemical Research Methods
Mohsin Ali Nasir, Samia Nawaz, Jian Huang
Summary: This article introduces the computational tools of CRISPR and their importance in gene editing. It also proposes new ideas and methods to improve existing computational tools and overcome their limitations.
CURRENT BIOINFORMATICS
(2022)
Article
Chemistry, Organic
Christian Fischer, Nynke A. Veprek, Zisis Peitsinis, Klaus-Peter Ruhmann, Chao Yang, Jessica N. Spradlin, Dustin Dovala, Daniel K. Nomura, Yingkai Zhang, Dirk Trauner
Summary: The COVID-19 pandemic has driven scientists to investigate potential remedies for SARS-CoV-2 and related viruses. Through virtual screening and molecular modeling, a class of easily accessible and quickly diversified small molecules have been identified as noncovalent inhibitors of the viral main protease. This highlights the potential for developing new treatments for coronaviruses in the future.
Article
Mathematical & Computational Biology
Anteneh M. Birga, Liping Ren, Huaichao Luo, Yang Zhang, Jian Huang
Summary: This study identified 87 candidate high-confidence risk genes (HRGs) associated with rheumatoid arthritis (RA) through integrated omics data. Analysis showed that these HRGs were significantly associated with different aspects of RA. Furthermore, drug repositioning prediction revealed potential targets and drugs for RA treatment. This study provides new insights into the pathogenesis of RA and has implications for therapeutic development.
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
(2022)
Review
Biochemical Research Methods
Birga A. Mengesha, Lin Ning, Jian Huang
Summary: This review study investigated the association between Human Leukocyte Antigen (HLA) HLA-DRB1 alleles and the risk of Rheumatoid Arthritis (RA). The results demonstrated that certain HLA-DRB1 alleles were significantly associated with an increased risk of RA, while others were potentially protective against the disease. This study provides important insights into the relationship between HLA-DRB1 and the risk of RA in different ethnic groups.
CURRENT BIOINFORMATICS
(2022)
Article
Chemistry, Medicinal
Chao Yang, Yingkai Zhang
Summary: In this study, the robustness and applicability of machine-learning scoring functions were further improved by expanding the training set, developing meaningful features, using a linear empirical scoring function as the baseline, and applying extreme gradient boosting (XGBoost) with Delta-machine learning. The new scoring function demonstrated superior performance in scoring and ranking in various structure types and showed reliability and robustness in virtual screening applications.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Genetics & Heredity
Yuwei Zhou, Shiyang Xie, Yue Yang, Lixu Jiang, Siqi Liu, Wei Li, Hamza Bukari Abagna, Lin Ning, Jian Huang
Summary: Therapeutic antibodies play a crucial role in the treatment of various diseases, but the success rate of antibody drug development is low, mainly due to the high aggregation tendency of antibody molecules. Therefore, developing efficient and high-throughput computational tools to predict the risk of hydrophobic interaction of antibodies is crucial.
FRONTIERS IN GENETICS
(2022)
Article
Genetics & Heredity
Xue Chen, Qianyue Zhang, Bowen Li, Chunying Lu, Shanshan Yang, Jinjin Long, Bifang He, Heng Chen, Jian Huang
Summary: The study focuses on the identification of blood-brain barrier penetrating peptides (BBPs) as drug candidates for central nervous system diseases, using a computational approach to quickly and accurately identify BBPs and non-BBPs. By creating training and testing datasets, the study found that the random forest method outperformed other classification algorithms in predicting BBPs. The newly developed predictor, BBPpredict, shows better performance compared to existing tools and can potentially contribute to the discovery of novel BBPs.
FRONTIERS IN GENETICS
(2022)
Review
Biochemistry & Molecular Biology
Chao Yang, Eric Anthony Chen, Yingkai Zhang
Summary: Molecular docking plays a significant role in early-stage drug discovery, and its success relies on the protein-ligand scoring function. This review provides an overview of recent scoring function development and docking-based applications in drug discovery. It discusses the strategies and resources for structure-based virtual screening, as well as the evaluation and development of classical and machine learning protein-ligand scoring functions. The review highlights the recent progress in machine learning scoring functions, including descriptor-based models and deep learning approaches. It also discusses the general workflow and docking protocols of structure-based virtual screening, along with a case study on large-scale docking-based virtual screening.
Article
Biochemistry & Molecular Biology
Jing Lin, Li Wen, Yuwei Zhou, Shaozhou Wang, Haiyang Ye, Jun Su, Juelin Li, Jianping Shu, Jian Huang, Peng Zhou
Summary: In this study, a comprehensive platform called PepQSAR database was developed, which systematically collects and decomposes various data sources and abundant information related to pQSARs. The database also includes a comparison function for previously built pQSAR models reported by different groups. This structured and searchable database is expected to be a useful resource and powerful tool for the computational peptidology community.
Article
Chemistry, Medicinal
Liping Ren, Xianrun Pan, Lin Ning, Di Gong, Jian Huang, Kejun Deng, Lei Xie, Yang Zhang
Summary: In this study, a liver cancer prognosis model was constructed using four hypoxia-related genes (NDRG1, ENO1, SERPINE1, ANXA2) identified from two independent datasets. The model showed significant differences in survival and immune characteristics between high- and low-risk groups, indicating its potential as a predictor and drug target for liver cancer prognosis. This study provides insights into the association between hypoxia, tumor microenvironment, and liver cancer prognosis.
CURRENT COMPUTER-AIDED DRUG DESIGN
(2023)
Article
Biochemistry & Molecular Biology
Mingyou Liu, Hongmei Liu, Tao Wu, Yingxue Zhu, Yuwei Zhou, Ziru Huang, Changcheng Xiang, Jian Huang
Summary: The ongoing COVID-19 pandemic necessitates the development of safe and efficient anti-coronavirus infection drugs. This study presents the ACP-Dnnel model, which employs machine learning techniques to predict anti-coronavirus peptides. The model achieves high accuracy and can expedite the discovery of anti-coronavirus peptide drugs.
Article
Biochemical Research Methods
Yuwei Zhou, Ziru Huang, Wenzhen Li, Jinyi Wei, Qianhu Jiang, Wei Yang, Jian Huang
Summary: Antibody drugs have become essential in biotherapeutics and have benefited patients with various diseases. However, their development process is time-consuming, costly, and risky. To accelerate development, reduce costs, and increase success rates, artificial intelligence, particularly deep learning methods, are extensively used in all stages of preclinical antibody drug development. This review systematically summarizes the use of deep learning in antibody drug discovery and development, including antibody encodings, deep learning architectures, and models. We also critically discuss the challenges, opportunities, current applications, and future directions of deep learning in antibody drug development.
Article
Multidisciplinary Sciences
Zhaohui Zhong, Guanqing Liu, Zhongjie Tang, Shuyue Xiang, Liang Yang, Lan Huang, Yao He, Tingting Fan, Shishi Liu, Xuelian Zheng, Tao Zhang, Yiping Qi, Jian Huang, Yong Zhang
Summary: In this study, a probiotic sourced CRISPR-LrCas9 system with a similar PAM requirement to Cas12a was reported, and its high efficiency in various genome editing applications was demonstrated.
NATURE COMMUNICATIONS
(2023)
Article
Mathematical & Computational Biology
Bowen Li, Heng Chen, Jian Huang, Bifang He
Summary: The CD47/SIRPa pathway is a new breakthrough in tumor immunity, and we developed a predictive model using NGPD and traditional machine learning methods to distinguish CD47 binding peptides. We screened CD47 binding peptides using NGPD biopanning technology and built computational models using multiple peptide descriptors and deep learning methods. The integrated model based on support vector machine showed good specificity, accuracy, and sensitivity during the cross-validation.
INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES
(2023)
Article
Biology
Lixu Jiang, Lin Ning, Chunchao Pu, Zixin Wang, Bifang He, Jian Huang
Summary: The research revealed that NgAgo in Natronobacterium gregoryi sp2 primarily binds to RNA, specifically transcripts of genes encoding tRNA, transcriptional regulators, RNA polymerases, and RNA-binding proteins. The findings suggest that NgAgo may play a role in post-transcriptional regulation in vivo.