Article
Biochemical Research Methods
Vlad-Rares Danaila, Catalin Buiu
Summary: This article presents a method based on neural networks and transfer learning to predict the sensitivity of HIV antibodies using the CATNAP dataset. Compared to other methods, this approach achieves better predictive performance and does not require structural features, making it more practical.
Article
Immunology
Allison Nau, Yun Shen, Vaishali Sanchorawala, Tatiana Prokaeva, Gareth J. Morgan
Summary: Monoclonal antibody light chain proteins cause tissue damage. By extracting complete light chain sequences from high throughput sequencing data, we identified a large number of multiple myeloma-associated monoclonal light chain sequences. This study is significant for understanding the pathogenic mechanisms of light chains.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Biochemical Research Methods
Andrzej Zielezinski, Sebastian Deorowicz, Adam Gudys
Summary: The Phage-Host Interaction Search Tool (PHIST) is a tool that predicts prokaryotic hosts of viruses based on exact matches between viral and host genomes. It improves host prediction accuracy at the species level and is significantly faster than other alignment-based tools, making it suitable for metagenomics studies.
Article
Chemistry, Medicinal
Ryann Perez, Xinning Li, Sam Giannakoulias, E. James Petersson
Summary: This article describes the use of the AggBERT approach, utilizing the ProtBERT model, to predict peptide amyloidogenesis. The results demonstrate that large language models have the potential to improve the accuracy and speed of amyloid fibril prediction in chemical biology and biomedicine.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Biochemistry & Molecular Biology
Tobias H. Olsen, Fergus Boyles, Charlotte M. Deane
Summary: The antibody repertoires are crucial for exploring disease states, vaccine responses, and therapeutic development, but accessing and processing data remains a challenge. The OAS database provides clean, annotated, and translated antibody repertoire data, with updates to accommodate increasing data volume and paired sequencing data. OAS now offers a new web server and standardized search parameters, making it a valuable resource.
Article
Multidisciplinary Sciences
Hooman H. Rashidi, Luke T. Dang, Samer Albahra, Resmi Ravindran, Imran H. Khan
Summary: MILO is an automated machine learning platform that efficiently integrates data processing, model training, and model validation to generate and evaluate numerous models for serological diagnosis of active tuberculosis. The 23-antigen model developed by MILO showed high sensitivity and specificity on secondary and tertiary datasets, demonstrating its potential for rapid clinical implementation in emerging infectious diseases.
SCIENTIFIC REPORTS
(2021)
Article
Biology
Md Wasi Ul Kabir, Duaa Mohammad Alawad, Avdesh Mishra, Md Tamjidul Hoque
Summary: This study developed an intelligent computer model called TAFPred to predict protein movements and torsion angle fluctuations based on their sequences. By analyzing various features of the protein sequences, the model achieved high accuracy in estimating protein flexibility. The use of the machine learning technique LightGBM further improved the predictions compared to existing methods. This study is important for understanding protein function and structure, and it has potential applications in medicine and biology.
Article
Medicine, Research & Experimental
Jingxian Zhou, Xuan Wang, Zhen Wei, Jia Meng, Daiyun Huang
Summary: This study presents a computational framework for predicting 4acC-carrying regions in Arabidopsis genomic DNA and demonstrates its promising performance through cross-validation and independent testing. The framework also enables motif mining and provides a user-friendly web server. The findings of this study will contribute to 4acC research.
MOLECULAR THERAPY-NUCLEIC ACIDS
(2022)
Article
Biochemical Research Methods
Wang Liu-Wei, Senay Kafkas, Jun Chen, Nicholas J. Dimonaco, Jesper Tegner, Robert Hoehndorf
Summary: This study developed DeepViral, a deep learning based method for predicting protein-protein interactions between humans and viruses. By combining learning from protein sequences and phenotype features, DeepViral significantly improves upon existing sequence-based methods.
Article
Biology
Mohamed E. M. Elhaj-Abdou, Hassan El-Dib, Amr El-Helw, Mohamed El-Habrouk
Summary: This paper introduces a hybrid deep neural network model, Deep_CNN_LSTM_GO, to predict unknown protein functions from sequences. The model integrates Convolutional Neural network (CNN) and Long Short-Term Memory (LSTM) Neural Network to learn features from amino acid sequences and outputs the three different Gene Ontology (GO) representing protein functions in Molecular Functions (MF), Biological Process (BP), and Cellular Component (CC). The proposed model outperforms other methods in the field with better performance using three evaluation metrics in the three sub-ontologies.
COMPUTATIONAL BIOLOGY AND CHEMISTRY
(2021)
Article
Multidisciplinary Sciences
Kotetsu Kayama, Miyuki Kanno, Naoto Chisaki, Misaki Tanaka, Reika Yao, Kiwamu Hanazono, Gerry Amor Camer, Daiji Endoh
Summary: A novel method was developed to predict the success of PCR amplification using a recurrent neural network, achieving 70% accuracy in predicting PCR results. This study suggests that neural networks could be used for primer design and PCR result prediction.
SCIENTIFIC REPORTS
(2021)
Article
Biochemical Research Methods
Jun Li, Sicheng Zhang, Dong Zhang, Shi-Jie Chen
Summary: RNA 3D structures play a crucial role in understanding their functions and designing drugs targeting RNA. However, experimentally determining RNA 3D structures is labor-intensive and technically challenging, resulting in a significant gap between the number of sequences and the availability of RNA structures. Therefore, computer-aided prediction of RNA 3D structures from sequences has become a highly desirable solution. In this study, a pipeline server integrating Vfold2D, Vfold3D, and VfoldLA programs is presented, enabling efficient and accurate prediction of RNA 3D structures or reliable initial structures for further refinement using an expanded 3D template database and 2D structural constraints extracted from the Rfam database.
Article
Biochemical Research Methods
Ye Yuan, Qushuo Chen, Jun Mao, Guipeng Li, Xiaoyong Pan
Summary: This study introduces a novel sequence-based antigen-antibody affinity prediction method, DG-Affinity, which utilizes deep neural networks to accurately predict the affinity between antibodies and antigens from sequences without structural information. DG-Affinity outperforms existing structure-based prediction methods and sequence-based tools, achieving a Pearson's correlation of over 0.65 on an independent test dataset, demonstrating its potential to advance antibody design.
BMC BIOINFORMATICS
(2023)
Article
Chemistry, Multidisciplinary
Sofie Nystrom, Per Hammarstrom
Summary: SARS-CoV-2 infection is associated with various morbidities. The spike protein of SARS-CoV-2 shows similarities to amyloid-disease associated blood coagulation and fibrinolytic disturbances, as well as neurologic and cardiac problems. Through experiments, researchers identified seven amyloidogenic sequences within the spike protein, and found that synthetic spike peptides can form amyloid-like fibrils when co-incubated with the protease neutrophil elastase. These findings suggest a potential molecular mechanism for amyloidogenesis of SARS-CoV-2 spike protein, which may be important in understanding COVID-19 disease-associated pathogenesis.
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
(2022)
Article
Multidisciplinary Sciences
Weam Fallatah, Ronika De, David Burks, Rajeev K. Azad, Pudur Jagadeeswaran
Summary: The zebrafish is a valuable model for studying thrombocyte function and development. This study used single-cell RNA sequencing to compare the gene expression profiles of young and mature thrombocytes in zebrafish. The results revealed similarities and differences between the two cell types, suggesting that thrombocytes have more features of megakaryocytes and that platelets may be considered as thrombocyte equivalents. This study provides insights into the process of thrombocyte maturation and lays the foundation for future research on megakaryocyte maturation.
Article
Oncology
Yao-Kuang Wang, Hao-Yi Syu, Yi-Hsun Chen, Chen-Shuan Chung, Yu Sheng Tseng, Shinn-Ying Ho, Chien-Wei Huang, I-Chen Wu, Hsiang-Chen Wang
Summary: The study developed a single-shot multibox detector using a convolutional neural network to diagnose esophageal neoplasms and evaluated its diagnostic accuracy, showing great potential of AI systems in identifying esophageal neoplasms and differentiating histological grades.
Article
Food Science & Technology
Chia-Chi Wang, Yu-Chih Liang, Shan-Shan Wang, Pinpin Lin, Chun-Wei Tung
Summary: This study proposes a machine learning-based weight-of-evidence (WoE) model for prioritizing chemicals of carcinogenic concern. The model integrates complementary computational methods and achieves better performance compared to single methods, providing a fast and comprehensive approach for prioritizing chemicals of carcinogenic concern.
FOOD AND CHEMICAL TOXICOLOGY
(2022)
Article
Multidisciplinary Sciences
Srinivasulu Yerukala Sathipati, Sanjay K. Shukla, Shinn-Ying Ho
Summary: This study proposes a spike protein predictor SPIKES, incorporating with a genetic algorithm, to determine the biochemical properties of spike proteins and their specificity to human hosts. The study identifies compositional differences at the amino acid sequence level between human and diverse animal coronaviruses, which may provide insights into the development and transmission of SARS-CoV-2 in humans and other species.
Article
Multidisciplinary Sciences
Srinivasulu Yerukala Sathipati, Ming-Ju Tsai, Sanjay K. Shukla, Shinn-Ying Ho, Yi Liu, Afshin Beheshti
Summary: This study successfully identified a miRNA signature and developed a survival estimation method for patients with bladder urothelial carcinoma (BLC). The miRNA signature can be used to estimate survival and also includes potential biomarkers for diagnosis and prognosis of BLC. This finding is important for understanding BLC and developing miRNA-targeted therapies.
SCIENTIFIC REPORTS
(2022)
Article
Environmental Sciences
Run-Hsin Lin, Chia-Chi Wang, Chun-Wei Tung
Summary: This study developed blood-sample gene biomarkers to predict stable MCI patients using feature selection and machine learning algorithms. Utilizing datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI), 29 gene biomarkers (31 probes) were identified for predicting stable MCI patients. A random forest-based classifier showed good performance with AUC values of 0.841 and 0.775 for cross-validation and test datasets, respectively, and achieved 97% concordance for patients with prediction score > 0.9.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Oncology
Yu-Kuei Chen, Ngo Tran My Ngoc, Hsi-Wen Chang, Ying-Fang Su, Chung-Hwan Chen, Yih-Gang Goan, Jeff Yi-Fu Chen, Chun-Wei Tung, Tzyh-Chyuan Hour
Summary: This study found that chlorogenic acid can inhibit the metastasis and invasion ability of esophageal squamous cell carcinoma by affecting the EGFR/p-Akt/Snail pathway.
ANTICANCER RESEARCH
(2022)
Article
Toxicology
Hung-Lin Kan, Chun-Wei Tung, Shao-En Chang, Ying-Chi Lin
Summary: Exposure to neurotoxicants has been associated with Parkinson's disease, but the identification of these neurotoxicants relies on animal models due to the clinical variation and slow progression of the disease. In this study, an innovative in silico model was proposed for predicting parkinsonian neurotoxicants. The model showed high specificity in ruling out non-neurotoxic chemicals and successfully predicted several chemicals related to parkinsonian motor deficits, providing a potential tool for prioritizing chemicals for further evaluations on Parkinson's disease potential.
ARCHIVES OF TOXICOLOGY
(2022)
Article
Medicine, Research & Experimental
Wen-Jane Lee, Keng-Hung Lin, Jun-Sing Wang, Wayne Huey-Herng Sheu, Chin-Chang Shen, Cheng-Ning Yang, Sheng-Mao Wu, Li-Wei Shen, Shu-Hua Lee, De-Wei Lai, Keng-Li Lan, Chun-Wei Tung, Shing-Hwa Liu, Meei-Ling Sheu
Summary: This study investigated the role of aryl hydrocarbon receptor (AhR) in diabetic retinopathy (DR) and found that the downregulation of AhR is associated with the progression of DR. The use of AhR agonists could reverse this downregulation and potentially maintain retinal vascular homeostasis, delaying the development of DR.
BIOMEDICINE & PHARMACOTHERAPY
(2022)
Article
Biochemical Research Methods
Che-Yu Chou, Pinpin Lin, Jongwoon Kim, Shan-Shan Wang, Chia-Chi Wang, Chun-Wei Tung
Summary: In this study, a valid prediction model for ex vivo human placental barrier permeability was developed, which can be used for assessing the toxic effects of chemicals on the fetus.
BMC BIOINFORMATICS
(2022)
Article
Medicine, Legal
Hui-Lun Lin, Yu-Wen Chiu, Chia-Chi Wang, Chun-Wei Tung
Summary: This study presents the development of a computational model for predicting in vitro pulmonary permeability of chemicals. By integrating multiple algorithms and applying applicability domain adjustment, the model achieved good performance in both cross-validation and independent testing.
REGULATORY TOXICOLOGY AND PHARMACOLOGY
(2022)
Article
Environmental Sciences
Shan-Shan Wang, Chia-Chi Wang, Chun-Wei Tung
Summary: Skin sensitization is an important endpoint associated with allergic contact dermatitis. Several alternative methods based on adverse outcome pathway (AOP) were developed to replace animal testing for evaluating skin sensitizers. An integrated testing strategy (ITS) that combines laboratory data and in silico methods has been proposed as a promising approach for hazard and potency assessment.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Entomology
Chia-Chi Wang, Shan-Shan Wang, Chun-Lin Liao, Wei-Ren Tsai, Chun-Wei Tung
Summary: Adverse outcome pathway (AOP)-based computational models are promising alternatives to animal testing, but their applicability in the field of pesticides needs further investigation. This study identified a consensus chemical space by comparing the predicted results with animal testing data, aiming to reduce animal testing for pesticides.
JOURNAL OF PESTICIDE SCIENCE
(2022)
Article
Food Science & Technology
Shan -Shan Wang, Pinpin Lin, Chia-Chi Wang, Ying-Chi Lin, Chun-Wei Tung
Summary: This study developed a nonlinear machine learning method to predict the migration of chemicals from packaging materials to food, taking into account chemical properties, material type, food type, and temperature. The ensemble model leveraging multiple algorithms provides better performance compared to previous linear regression models. These prediction models can accelerate the assessment of migration of food contact chemicals from package to foods.
FOOD AND CHEMICAL TOXICOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Wan-Chi Hsiao, Guang-Hao Niu, Chen-Fu Lo, Jing-Ya Wang, Ya-Hui Chi, Wei-Cheng Huang, Chun-Wei Tung, Ping-Jyun Sung, Lun Kelvin Tsou, Mingzi M. Zhang
Summary: In this study, two covalent probes were developed to identify STING as a direct target of excB in living mammalian cells. The study reveals a possible mechanism-of-action of excB and expands the repertoire of covalent STING inhibitors.
COMMUNICATIONS CHEMISTRY
(2023)
Article
Veterinary Sciences
Chia-Chi Wang, Yu-Ting Hung, Che-Yu Chou, Shih-Ling Hsuan, Zeng-Weng Chen, Pei-Yu Chang, Tong-Rong Jan, Chun-Wei Tung
Summary: Antimicrobial resistance is a global health issue and surveillance of AMR can provide valuable insights. Salmonella, widely found in food-producing animals, is prioritized in the AMR surveillance program by WHO. The study aims to develop WGS-based random forest models for predicting MIC values of 24 drugs using data from labs in Taiwan.
VETERINARY RESEARCH
(2023)