Review
Biochemical Research Methods
Maryam Bagherian, Elyas Sabeti, Kai Wang, Maureen A. Sartor, Zaneta Nikolovska-Coleska, Kayvan Najarian
Summary: Predicting interactions between drugs and targets is crucial in drug discovery, necessitating the development of efficient prediction approaches to avoid costly and uncertain experimental determinations. These approaches should timely identify potential drug-target interactions.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Biochemistry & Molecular Biology
Jinyin Zha, Mingyu Li, Ren Kong, Shaoyong Lu, Jian Zhang
Summary: Allostery is an important regulatory phenomenon in life processes and disease therapy, but studying it has been challenging due to the lack of knowledge. To address this, we created the Allosteric Database (ASD) and reviewed the four categories of data in this database and how researchers have utilized them for their studies. Several new drug targets and allosteric modulators discovered through the use of ASD are also highlighted.
JOURNAL OF MOLECULAR BIOLOGY
(2022)
Article
Biochemistry & Molecular Biology
Qihang Cai, Rongao Yuan, Jian He, Menglong Li, Yanzhi Guo
Summary: The study analyzed 21 drug resistances caused by mutated residues based on HIV target protein sequence information using machine learning methods, with the application of PCA to reduce feature dimensionality. Results indicated that the RBF-based SVM method showed superior performance in prediction, especially when incorporating weighted information to improve predictive ability.
MOLECULAR DIVERSITY
(2021)
Article
Biochemistry & Molecular Biology
Liwei Liu, Qi Zhang, Yuxiao Wei, Qi Zhao, Bo Liao
Summary: BG-DTI is a learning-based framework for predicting drug-target interactions. It combines approaches based on biological features and heterogeneous networks and utilizes a graph representation learning module to learn the features representation of drugs and targets. The fusion descriptors obtained from the module are fed into a random forest classifier for DTI prediction. Evaluation results demonstrate that BG-DTI outperforms other methods.
Review
Biochemical Research Methods
Lianlian Wu, Yuqi Wen, Dongjin Leng, Qinglong Zhang, Chong Dai, Zhongming Wang, Ziqi Liu, Bowei Yan, Yixin Zhang, Jing Wang, Song He, Xiaochen Bo
Summary: This article introduces the recent applications of machine learning in drug combination prediction and discusses the related databases and tools. It focuses on the concept and controversy of synergism between drug combinations and analyzes the challenges of ML methods in prediction tasks.
BRIEFINGS IN BIOINFORMATICS
(2022)
Review
Pharmacology & Pharmacy
Yaojia Chen, Liran Juan, Xiao Lv, Lei Shi
Summary: This article reviews the current research status of modeling-based anti-cancer drug sensitivity prediction, emphasizing the importance of genomics data in the prediction task, while also pointing out that existing prediction models neglect the significant impacts of gene mutations, methylation, and copy number variations on drug sensitivity.
FRONTIERS IN PHARMACOLOGY
(2021)
Article
Multidisciplinary Sciences
Matthew McPartlon, Jinbo Xu
Summary: Protein side-chain packing is important for predicting, refining, and designing protein structures. Existing methods for this task are not satisfactory in terms of speed and accuracy. AttnPacker is a deep learning method that directly predicts protein side-chain coordinates, incorporating backbone 3D geometry to improve computational efficiency. It produces physically realistic side-chain conformations, reducing steric clashes and improving accuracy compared to state-of-the-art methods.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Article
Computer Science, Information Systems
Arjun Puri, Manoj Kumar Gupta, Kanica Sachdev
Summary: This article proposes a model for studying drug-target interaction problems using computational techniques. The model uses feature representations and resampling techniques to handle class imbalance, and utilizes a soft voting ensemble method to improve prediction accuracy. Experiments demonstrate that the proposed model outperforms existing models on standard datasets.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Biochemical Research Methods
Mehdi Yazdani-Jahromi, Niloofar Yousefi, Aida Tayebi, Elayaraja Kolanthai, Craig J. Neal, Sudipta Seal, Ozlem Ozmen Garibay
Summary: In this study, an interpretable graph-based deep learning prediction model called AttentionSiteDTI is introduced to address the problem of drug-target interaction prediction. The model utilizes protein binding sites and a self-attention mechanism to identify the most contributive binding sites. Experimental results show that AttentionSiteDTI outperforms current state-of-the-art models on three benchmark datasets and exhibits high generalizability on new proteins. The agreement between computationally predicted and experimentally observed drug-target interactions demonstrates the potential of the proposed method as an effective pre-screening tool in drug repurposing applications.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Plant Sciences
Jonathan Wei Xiong Ng, Swee Kwang Chua, Marek Mutwil
Summary: Understanding how different cellular components work together to form a living cell is possible through multidisciplinary approaches combining molecular and computational biology. Machine learning has great potential in life sciences as it can discover novel relationships between biological features. Researchers created a dataset of gene features and developed a machine learning workflow to identify linked features. The detected linked features are visualized as a Feature Important Network (FIN), which provides insights into gene function. To enhance accessibility, the FINder database is made available to the scientific community.
FRONTIERS IN PLANT SCIENCE
(2022)
Review
Chemistry, Multidisciplinary
Maciej Staszak, Katarzyna Staszak, Karolina Wieszczycka, Anna Bajek, Krzysztof Roszkowski, Bartosz Tylkowski
Summary: The paper provides a comprehensive overview of the use of artificial intelligence systems, particularly neural networks, in drug design to identify chemical structures with medical relevance. Successful training of neural networks requires a large set of training data related to the chemical structure-biological activity relationship, which can be obtained from experimental measurements or appropriate quantum models. Recent advancements in computing power have led to rapid development in neural network systems and a growing interest in deep learning techniques, allowing for a new level of abstraction in network modeling.
WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE
(2022)
Review
Biochemistry & Molecular Biology
Zihao Chen, Long Hu, Bao-Ting Zhang, Aiping Lu, Yaofeng Wang, Yuanyuan Yu, Ge Zhang
Summary: Aptamers, short nucleic acid molecules, show promise as antibody alternatives for diagnostics and therapeutics due to their unique features. The SELEX process for aptamer selection is time-consuming, calling for artificial intelligence assistance in candidate identification. Machine/deep learning methods offer potential for predicting aptamer-target binding in a more efficient manner.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2021)
Article
Mathematical & Computational Biology
Sofia D'Souza, K. V. Prema, S. Balaji, Ronak Shah
Summary: Chemogenomics, or proteochemometrics, uses computational methods to predict drug-target interactions based on large-scale data. This study develops a deep learning CNN model using one-dimensional SMILES for drugs and protein binding pocket sequences as inputs to predict unknown ligand-target interactions. The proposed method outperforms shallow machine learning methods in terms of prediction accuracy and computational efficiency.
INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES
(2023)
Article
Chemistry, Multidisciplinary
Guannan Liu, Manali Singha, Limeng Pu, Prasanga Neupane, Joseph Feinstein, Hsiao-Chun Wu, J. Ramanujam, Michal Brylinski
Summary: GraphDTI is a robust machine learning framework that integrates information on drugs, proteins, and binding sites with gene expression and protein-protein interactions, demonstrating high performance and generalizability for identifying drug targets. Applications of GraphDTI include investigating polypharmacological effects, side effects caused by off-target binding, and repositioning opportunities for drugs.
JOURNAL OF CHEMINFORMATICS
(2021)
Article
Biochemical Research Methods
Giulia Muzio, Leslie O'Bray, Karsten Borgwardt
Summary: Recent advancements in experimental high-throughput technologies have expanded the availability and quantity of molecular data in biology, creating a need for new computational tools to analyze networks. Graph neural networks (GNNs) are being frequently applied in bioinformatics for tasks such as protein function prediction, protein-protein interaction prediction, and in silico drug discovery and development. Deep learning is emerging as a new tool to answer classic questions in areas like gene regulatory networks and disease diagnosis.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Biochemical Research Methods
Tianyi Zhao, Yang Hu, Liang Cheng
Summary: This study introduces a graph deep learning method, Deep-DRM, for identifying diseases-related metabolites. By calculating the similarities between metabolites and diseases, building networks, and applying a deep neural network, Deep-DRM shows outstanding performance in identifying true metabolite-disease pairs.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Biochemical Research Methods
Xudong Han, Qingfei Kong, Chonghui Liu, Liang Cheng, Junwei Han
Summary: SubtypeDrug is a software package based on systems biology that prioritizes subtype-specific drugs based on cancer expression data from samples of multiple subtypes. It uses a novel approach to identify subtype-specific drugs by considering the biological functions regulated by drugs at the subpathway level. Its capabilities include extraction of subpathways from biological pathways, identification of dysregulated subpathways induced by each drug, inference of sample-specific subpathway activity profiles, evaluation of drug-disease reverse association at the subpathway level, identification of cancer-subtype-specific drugs through subtype sample set enrichment analysis, and visualization of the results.
Article
Biochemical Research Methods
Yuqi Sheng, Ying Jiang, Yang Yang, Xiangmei Li, Jiayue Qiu, Jiashuo Wu, Liang Cheng, Junwei Han
Summary: Biological pathways are crucial in dictating disease states and drug responses, dysfunctional subpathways (SPs) have been associated with cancer, high-throughput sequencing allows for identification of dysfunctional SPs, a novel network-based method, CNA2Subpathway, integrates pathway topology information, multi-omics data, and SP crosstalk to identify cancer-relevant SPs driven by somatic CNAs. Validated in breast cancer and head and neck cancer datasets, CNA2Subpathway shows effectiveness in uncovering dysfunctional SPs associated with cancer immune response and patient prognosis.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Multidisciplinary Sciences
Li Chen, Hong-Ming Zhu, Yan Li, Qi-Fa Liu, Yu Hu, Jian-Feng Zhou, Jie Jin, Jian-Da Hu, Ting Liu, De-Pei Wu, Jie-Ping Chen, Yong-Rong Lai, Jian-Xiang Wang, Juan Li, Jian-Yong Li, Xin Du, Xin Wang, Ming-Zhen Yang, Jin-Song Yan, Gui-Fang Ouyang, Li Liu, Ming Hou, Xiao-Jun Huang, Xiao-Jing Yan, Dan Xu, Wei-Ming Li, Deng-Ju Li, Yin-Jun Lou, Zheng-Jun Wu, Ting Niu, Ying Wang, Xiao-Yang Li, Jian-Hua You, Hui-Jin Zhao, Yu Chen, Yang Shen, Qiu-Sheng Chen, Jian Li, Bing-Shun Wang, Wei-Li Zhao, Jian-Qing Mi, Kan-Kan Wang, Jiong Hu, Zhu Chen, Sai-Juan Chen, Jun-Min Li
Summary: This study found that the combination of all-trans retinoic acid and arsenic trioxide in treating acute promyelocytic leukemia during consolidation therapy is not inferior to traditional chemotherapy regimens and shows better outcomes in reducing relapse and toxicity.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Genetics & Heredity
Zijun Zhu, Xudong Han, Liang Cheng
Summary: This study integrates genetic mutation information and expression data to obtain a predictive signature for T2DM, which shows excellent performance and robustness.
CURRENT GENE THERAPY
(2022)
Review
Biochemistry & Molecular Biology
Haixiu Yang, Changlu Qi, Boyan Li, Liang Cheng
Summary: Recent studies have revealed the significant role of non-coding RNAs in regulating drug resistance in tumor cells. Understanding the relationship between ncRNAs and drug resistance can provide valuable insights into the mechanisms and biomarkers of chemoresistance, as well as facilitate personalized anticancer treatment regimens.
CURRENT MEDICINAL CHEMISTRY
(2022)
Letter
Infectious Diseases
Shizheng Qiu, Yang Hu, Liang Cheng
JOURNAL OF INFECTION
(2022)
Article
Biochemistry & Molecular Biology
Liang Cheng, Changlu Qi, Haixiu Yang, Minke Lu, Yiting Cai, Tongze Fu, Jialiang Ren, Qu Jin, Xue Zhang
Summary: gutMGene, a manually curated database, provides a comprehensive resource of target genes of gut microbes and microbial metabolites in humans and mice. The database documents curated relationships between gut microbes, microbial metabolites and genes in both species. The user-friendly interface allows browsing and retrieval of entries, and also offers the option to download all entries and submit new validated associations.
NUCLEIC ACIDS RESEARCH
(2022)
Review
Biochemical Research Methods
Zijun Zhu, Sainan Zhang, Ping Wang, Xinyu Chen, Jianxing Bi, Liang Cheng, Xue Zhang
Summary: This article reviews the advantages and limitations of integrating omics data for SARS-CoV-2 and COVID-19, and summarizes the existing data analysis methods and research results. It also proposes research directions for the integration of SARS-CoV-2 (COVID-19) multi-omics data and presents a case study for deeper exploration of the disease mechanisms of COVID-19.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemistry & Molecular Biology
Changlu Qi, Chao Wang, Lingling Zhao, Zijun Zhu, Ping Wang, Sainan Zhang, Liang Cheng, Xue Zhang
Summary: SCovid is a comprehensive resource of single-cell data to explore the molecular characteristics of COVID-19 in different human tissues. It collects 21 single-cell datasets and reveals the molecular features of COVID-19 through manual annotation.
NUCLEIC ACIDS RESEARCH
(2022)
Letter
Multidisciplinary Sciences
Zijun Zhu, Xinyu Chen, Chao Wang, Liang Cheng
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Article
Biochemical Research Methods
Junjie Wang, Jie Hu, Huiting Sun, MengDie Xu, Yun Yu, Yun Liu, Liang Cheng
Summary: In this study, we propose a multigranularity protein-ligand interaction model, which utilizes the Transformer model and convolutional neural network to accurately predict the binding affinity between drugs and protein targets.
Article
Biochemistry & Molecular Biology
Changlu Qi, Yiting Cai, Kai Qian, Xuefeng Li, Jialiang Ren, Ping Wang, Tongze Fu, Tianyi Zhao, Liang Cheng, Lei Shi, Xue Zhang
Summary: The gut microbiota plays a crucial role in maintaining health, and disruptions can lead to disorders. The gutMDisorder database provides a valuable resource for studying dysbiosis, and the latest version offers expanded data and improved features.
NUCLEIC ACIDS RESEARCH
(2023)
Letter
Gastroenterology & Hepatology
Zijun Zhu, Xinyu Chen, Chao Wang, Sainan Zhang, Liang Cheng
Article
Virology
Zijun Zhu, Xinyu Chen, Chao Wang, Sainan Zhang, Rui Yu, Yubin Xie, Shuofeng Yuan, Liang Cheng, Lei Shi, Xue Zhang
Summary: This study conducted a genome-wide association study (GWAS) to identify host genetic factors associated with COVID-19. The correlation between genetic variations and gene expression was assessed using expression quantitative trait locus (eQTL) analysis. The findings revealed 20 genes significantly associated with immunity and neurological disorders. Single-cell datasets were used to validate these findings and to explore the causal relationship between COVID-19 and neurological disorders. Cell experiments were conducted to investigate the effects of COVID-19-related protein-coding genes. This study provides important insights into the genetic architecture underlying the pathophysiology of COVID-19.
JOURNAL OF MEDICAL VIROLOGY
(2023)