Article
Computer Science, Information Systems
Jiayun Chang, Xiongjun Fu, Congxia Zhao, Ping Lang, Cheng Feng
Summary: This paper proposes an improved RF-SSA algorithm to balance detection efficiency and accuracy. The algorithm combines the 3DT-WD method and SSA algorithm to obtain higher detection performance through parameter optimization. Simulation experiments and real data results have demonstrated the superiority of this approach.
Article
Biochemistry & Molecular Biology
Jiali Song, Zhenyi Xu, Lei Cao, Meng Wang, Yan Hou, Kang Li
Summary: The study developed a feature-based method, PsePDC-DTIs, to predict drug-target interactions (DTIs) for breast cancer research, achieving good prediction results and exploring new treatment targets for breast cancer.
Article
Biochemistry & Molecular Biology
Shimei Qin, Wan Li, Hongzheng Yu, Manyi Xu, Chao Li, Lei Fu, Shibin Sun, Yuehan He, Junjie Lv, Weiming He, Lina Chen
Summary: Drug repositioning is an effective approach to develop drugs for complex diseases like cancer and network-based computational biology approaches have been successfully applied to drug repurposing. In this study, a new strategy for network-based drug repositioning against cancer was developed. By constructing a cancer-related drug similarity network and quantifying the correlation score of each drug with specific cancer, potential repositionable drugs were identified and confirmed by clinical trial articles and databases. The targets of these drugs were significantly associated with the prognosis of NSCLC and provided valuable perspective for drug repurposing in cancer.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Review
Pharmacology & Pharmacy
Jun-Lin Yu, Qing-Qing Dai, Guo-Bo Li
Summary: Drug repositioning is an attractive strategy for discovering new therapeutic uses for approved or investigational drugs. Deep learning has gained attention for its potential in target prediction and drug repositioning, improving efficiency and success rates.
DRUG DISCOVERY TODAY
(2022)
Review
Biochemistry & Molecular Biology
Karim Abbasi, Parvin Razzaghi, Antti Poso, Saber Ghanbari-Ara, Ali Masoudi-Nejad
Summary: This study provides a comprehensive overview of deep learning-based drug-target interactions (DTIs) prediction approaches, exploring different deep network architectures, commonly used datasets, and current challenges in the field. The research findings reveal the potential of deep learning in DTIs prediction and suggest future directions for development.
CURRENT MEDICINAL CHEMISTRY
(2021)
Article
Biochemical Research Methods
Mengting Niu, Quan Zou
Summary: Single-guide RNA (sgRNA) is a non-coding RNA that guides the insertion or deletion of uridine residues into kinetoplastid during RNA editing. In this paper, a new classifier called SgRNA-RF is developed, which extracts features of nucleic acid composition and structure from the on-target activity sgRNA sequence and identifies them using the random forest algorithm. The classifier significantly improves the identification accuracy and provides a user-friendly web server for implementation.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Pharmacology & Pharmacy
Rui Xuan Huang, Damrongrat Siriwanna, William C. Cho, Tsz Kin Wan, Yan Rong Du, Adam N. Bennett, Qian Echo He, Jun Dong Liu, Xiao Tai Huang, Kei Hang Katie Chan
Summary: In this study, a pipeline based on machine learning was developed to predict potential target genes for LUAD and discover potential drugs for its treatment through drug repositioning. The pipeline achieved good predictive performance and identified several potential therapeutic drugs for LUAD.
FRONTIERS IN PHARMACOLOGY
(2022)
Article
Microbiology
Yu Peng, Shouwei Zhao, Zhiliang Zeng, Xiang Hu, Zhixiang Yin
Summary: Prediction of drug-target interactions (DTIs) is crucial for drug development. Traditional laboratory methods are time-consuming and expensive. Recent studies have shown that using machine learning methods, such as LightGBM, can accelerate the drug development process and reduce costs. In this study, we compared the performance of our new model, LGBMDF, with other state-of-the-art methods using cross-validation, and found that our method has better accuracy and faster computation.
FRONTIERS IN MICROBIOLOGY
(2023)
Article
Biochemical Research Methods
Lu Jiang, Jiahao Sun, Yue Wang, Qiao Ning, Na Luo, Minghao Yin
Summary: Accurate identification of drug-target interactions is crucial in drug discovery. This paper proposes a new method, DTIHNC, that integrates heterogeneous networks and cross-modal similarities to identify drug-target interactions. The method outperforms state-of-the-art methods and demonstrates its effectiveness.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Automation & Control Systems
S. M. Hasan Mahmud, Wenyu Chen, Hosney Jahan, Yougsheng Liu, S. M. Mamun Hasan
Summary: The study highlights the significance of predicting novel drug-target interactions (DTIs) and proposes a new multi-label approach, idti-MLKdr, using multi-kernel learning (MKL) based SVM for DTI prediction, which shows superior performance. By extracting drug-target features, constructing negative drug-target pairs, applying dimensionality reduction techniques, the method achieves promising results, motivating further drug development research.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2021)
Article
Immunology
Peng Han, Chao Hou, Xi Zheng, Lulu Cao, Xiaomeng Shi, Xiaohui Zhang, Hua Ye, Hudan Pan, Liang Liu, Tingting Li, Fanlei Hu, Zhanguo Li
Summary: This study investigated the serum antigenomic profiling in rheumatoid arthritis (RA) using label-free proteomic technology and machine-learning algorithm. The results identified potential diagnostic biomarkers and developed a model to classify different types of RA patients.
FRONTIERS IN IMMUNOLOGY
(2022)
Article
Biochemistry & Molecular Biology
Laura Ortiz-Miravalles, Manuel Sanchez-Angulo, Jesus M. Sanz, Beatriz Maestro
Summary: A collection of 1200 compounds from a repurposing drugs library was screened for their antimicrobial effects against Streptococcus pneumoniae. After four rounds of screening, seven compounds were selected and found to inhibit pneumococcal growth and reduce bacterial viability by 90.0% to 99.9% at a concentration of 25 mu M. All compounds, except one, also increased bacterial membrane permeability and shared a common chemical structure. These findings offer new possibilities for combating pneumococcal diseases through drug repositioning and provide insights for the design of membrane-targeted antimicrobials.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Biochemical Research Methods
Xian-rui Wang, Ting-ting Cao, Cong Min Jia, Xue-mei Tian, Yun Wang
Summary: This paper screened molecular descriptors based on molecular vibrations and considered molecule-target as a whole system to construct prediction models with high accuracy and wide applicability for drug-target interactions (DTIs) affinity, providing reference for quantifying DTIs affinity. The Random Forest (RF) models developed in this study showed higher accuracy and wider applicability compared to existing quantitative models, with importance of E-state molecular descriptors and certain protein descriptors identified in the quantification of DTIs.
BMC BIOINFORMATICS
(2021)
Article
Environmental Sciences
Mohammad Amin Hariri-Ardebili, Golsa Mahdavi, Azam Abdollahi, Ali Amini
Summary: A hybrid Random Field-Polynomial Chaos Expansion surrogate model for uncertainty quantification and sensitivity assessment of dams is proposed in this study. The most sensitive spatial locations within dam body for different vibration modes are identified using Sobol's indices and correlation rank methods. The results of the hybrid model are validated using the classical random forest regression method, improving system identification and dynamic analysis outcomes.
Article
Genetics & Heredity
He-Gang Chen, Xiong-Hui Zhou
Summary: Drug repurposing/repositioning using gene expression data and protein-protein interactions can lead to novel drug indications. The MNBDR method, combining random walk algorithms and a new indicator, is effective in identifying potential drugs and revealing biological mechanisms in drug response.
Article
Chemistry, Multidisciplinary
Xiaoqin Pan, Xuan Lin, Dongsheng Cao, Xiangxiang Zeng, Philip S. Yu, Lifang He, Ruth Nussinov, Feixiong Cheng
Summary: This review introduces guidelines on utilizing deep learning methodologies and tools for drug repurposing, which is of great importance in drug development. The article summarizes the commonly used bioinformatics and pharmacogenomics databases for drug repurposing and discusses the recently developed sequence-based and graph-based representation approaches as well as state-of-the-art deep learning-based methods. The applications of drug repurposing in fighting the COVID-19 pandemic are presented, along with an outline of future challenges.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE
(2022)
Article
Biochemical Research Methods
Zhenxing Wu, Dejun Jiang, Jike Wang, Xujun Zhang, Hongyan Du, Lurong Pan, Chang-Yu Hsieh, Dongsheng Cao, Tingjun Hou
Summary: Molecular property prediction models based on machine learning algorithms are important tools in early stages of drug discovery. This study introduces a new pre-training method, K-BERT, which can extract chemical information from SMILES and provides superior predictions compared to traditional models.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Xiao-Chen Zhang, Jia-Cai Yi, Guo-Ping Yang, Cheng-Kun Wu, Ting-Jun Hou, Dong-Sheng Cao
Summary: This paper presents a deep neural network model called ABC-Net, which can directly predict graph structures. By using the divide-and-conquer principle, atoms or bonds are modeled as single points in the center, and a fully convolutional neural network is leveraged to identify and predict relevant properties, enabling the recovery of molecular structures. Experimental results demonstrate significant improvement in recognition performance with this method.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Zhe Wang, Hong Pan, Huiyong Sun, Yu Kang, Huanxiang Liu, Dongsheng Cao, Tingjun Hou
Summary: The fastDRH server is a free and open accessed web platform for predicting and analyzing protein-ligand complex structures. It integrates multiple features such as molecular docking, docking pose rescoring, and hotspot residue prediction to provide key information to users clearly. With a success rate of >80% in benchmark for protein-ligand binding mode prediction, the fastDRH server is a reliable tool for drug discovery projects.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Chemistry, Applied
Zheng-Fei Yang, Ran Xiao, Guo-Li Xiong, Qin-Lu Lin, Ying Liang, Wen-Bin Zeng, Jie Dong, Dong-sheng Cao
Summary: A novel multi-layer sweetness evaluation system based on machine learning methods was proposed to evaluate sweet properties of compounds with different chemical spaces and categories, providing quantitative predictions of sweetness. The study obtained sweetness-related chemical basis and structure transforming rules using molecular cloud and matched molecular pair analysis (MMPA) methods. The research aims to facilitate food scientists with efficient screening and precise development of high-quality sweeteners.
Article
Chemistry, Medicinal
Xujun Zhang, Chao Shen, Ben Liao, Dejun Jiang, Jike Wang, Zhenxing Wu, Hongyan Du, Tianyue Wang, Wenbo Huo, Lei Xu, Dongsheng Cao, Chang-Yu Hsieh, Tingjun Hou
Summary: The development of accurate machine-learning-based scoring functions for virtual screening requires unbiased and diverse datasets. However, most existing datasets may suffer from hidden biases and data insufficiency. In this study, we developed a new approach named TocoDecoy to generate unbiased and expandable datasets, and evaluated its performance compared to other datasets.
JOURNAL OF MEDICINAL CHEMISTRY
(2022)
Article
Chemistry, Multidisciplinary
Ning-Ning Wang, Xiang-Gui Wang, Guo-Li Xiong, Zi-Yi Yang, Ai-Ping Lu, Xiang Chen, Shao Liu, Ting-Jun Hou, Dong-Sheng Cao
Summary: Drug-drug interactions (DDIs) can cause serious adverse reactions, making it necessary to develop effective in silico methods to predict and evaluate DDIs. In this study, high-performance predictive models for metabolic DDIs were constructed and validated, yielding more reliable predictions. The developed model will greatly contribute to future drug development and clinical pharmacy research.
JOURNAL OF CHEMINFORMATICS
(2022)
Article
Pharmacology & Pharmacy
Lingjie Bao, Zhe Wang, Zhenxing Wu, Hao Luo, Jiahui Yu, Yu Kang, Dongsheng Cao, Tingjun Hou
Summary: In this study, a model called AMGU was developed to predict the inhibitory activities of small molecules against various kinases. The AMGU model outperformed other models on both internal and external test sets, demonstrating its enhanced generalizability. Additionally, a method called edges masking was devised to explain the predictive mechanisms, and a web server called KIP was developed for predicting the polypharmacology effects of small molecules on the kinome.
ACTA PHARMACEUTICA SINICA B
(2023)
Article
Biochemistry & Molecular Biology
Gaoqi Weng, Xuanyan Cai, Dongsheng Cao, Hongyan Du, Chao Shen, Yafeng Deng, Qiaojun He, Bo Yang, Dan Li, Tingjun Hou
Summary: PROTAC-DB 2.0 is an updated online database that contains structural and experimental data about PROTACs. This second version expands the number of PROTACs to 3270 and provides additional information to aid in the understanding and design of PROTACs.
NUCLEIC ACIDS RESEARCH
(2023)
Article
Chemistry, Medicinal
Teng-Zhi Long, Shao-Hua Shi, Shao Liu, Ai-Ping Lu, Zhao-Qian Liu, Min Li, Ting-Jun Hou, Dong-Sheng Cao
Summary: This study constructed a high-quality dataset and established a series of classification models using machine learning algorithms to predict hematotoxicity. The best model based on Attentive FP showed excellent performance on both the validation and test sets. Additionally, the study utilized SHAP and atom heatmap methods to identify important features and structural fragments related to hematotoxicity, and employed MMPA and representative substructure derivation technique to further investigate the transformation principles and distinctive structural features of hematotoxic chemicals.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Multidisciplinary Sciences
Ruoru Wu, Zhihao Shu, Fei Zou, Shaoli Zhao, Saolai Chan, Yaxian Hu, Hong Xiang, Shuhua Chen, Li Fu, Dongsheng Cao, Hongwei Lu
Summary: In this study, the significance of serum myoglobin (Mb) in the pathogenesis of diabetic kidney disease (DKD) was investigated. The results suggest that serum Mb may serve as a potential indicator for DKD and is associated with renal function impairment caused by metabolic syndrome components.
SCIENTIFIC REPORTS
(2022)
Article
Instruments & Instrumentation
Jiayin Deng, Zhuyifan Ye, Wenwen Zheng, Jian Chen, Haoshi Gao, Zheng Wu, Ging Chan, Yongjun Wang, Dongsheng Cao, Yanqing Wang, Simon Ming-Yuen Lee, Defang Ouyang
Summary: Microspheres have attracted attention from the pharmaceutical and medical industry due to their excellent biodegradability and long controlled-release characteristics. This research successfully built a prediction model using machine learning techniques to accelerate microspheres product development for small-molecule drugs. The consensus model achieved high accuracy in predicting the in vitro drug release profiles and can provide meaningful insights for microspheres development.
DRUG DELIVERY AND TRANSLATIONAL RESEARCH
(2023)
Article
Biochemistry & Molecular Biology
Jiashun Mao, Shenghui Guan, Yongqing Chen, Amir Zeb, Qingxiang Sun, Ranlan Lu, Jie Dong, Jianmin Wang, Dongsheng Cao
Summary: Antimicrobial resistance could be a serious threat to millions of lives. Antimicrobial peptides (AMPs) offer an alternative to conventional antibiotics for combating infectious diseases. However, developing and optimizing AMPs face significant challenges, and advanced methods are needed to overcome these challenges and create effective AMP-driven treatments.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Multidisciplinary Sciences
Xiao-Chen Zhang, Cheng-Kun Wu, Jia-Cai Yi, Xiang-Xiang Zeng, Can-Qun Yang, Ai-Ping Lu, Hou T-j, Ting-Jun Hou, Dong-Sheng Cao
Summary: Accurate prediction of pharmacological properties of small molecules is crucial in drug discovery. However, existing deep learning methods face challenges in handling scarcity of labeled data and information sharing among different tasks. In this study, we propose a novel multitask learning BERT framework, MTL-BERT, which leverages large-scale pre-training, multitask learning, and SMILES enumeration to address the data scarcity problem. Experimental results show that MTL-BERT outperforms state-of-the-art methods on 60 practical molecular datasets and leverages attention mechanisms for model interpretability.
Article
Chemistry, Medicinal
Jike Wang, Xiaorui Wang, Huiyong Sun, Mingyang Wang, Yundian Zeng, Dejun Jiang, Zhenxing Wu, Zeyi Liu, Ben Liao, Xiaojun Yao, Chang-Yu Hsieh, Dongsheng Cao, Xi Chen, Tingjun Hou
Summary: In this study, a new algorithm called ChemistGA was proposed, which combines traditional heuristic algorithms with deep learning to design novel molecules. ChemistGA not only retains the strengths of traditional algorithms, but also greatly enhances the synthetic accessibility and success rate of generated molecules.
JOURNAL OF MEDICINAL CHEMISTRY
(2022)