Article
Public, Environmental & Occupational Health
Akshaya Karthikeyan, Akshit Garg, P. K. Vinod, U. Deva Priyakumar
Summary: This study introduces machine learning methods based on blood test data for predicting COVID-19 mortality risk, achieving high accuracy. By analyzing key biomarkers, it provides decision-making solutions for healthcare systems to expedite the decision-making process.
FRONTIERS IN PUBLIC HEALTH
(2021)
Article
Oncology
Bin Wang, Shan Zhang, Xubin Wu, Ying Li, Yueming Yan, Lili Liu, Jie Xiang, Dandan Li, Ting Yan
Summary: The radiomics models constructed based on multiparametric MRI data showed promising predictive accuracy for individualized estimation of survival stratification in GBM patients. The models differentiated between long-term and short-term survival groups with favorable discrimination ability, indicating their potential clinical value. Different MRI modalities and tumor subregions had varied effects on the survival indicators, with C indexes of up to 0.725, 0.677, and 0.724 achieved in the validation set.
FRONTIERS IN ONCOLOGY
(2021)
Article
Oncology
Shuchita Dhwiren Patel, Andrew Davies, Emma Laing, Huihai Wu, Jeewaka Mendis, Derk-Jan Dijk
Summary: Combining sleep-wake parameters with routine clinical data can improve survival prediction in advanced cancer patients. Machine learning models identified both established and new predictors of survival. This study is significant for the development of prognostic tools.
Review
Biochemistry & Molecular Biology
Thi Tuyet Van Tran, Hilal Tayara, Kil To Chong
Summary: Drug distribution is a crucial process in pharmacokinetics, as it affects the effectiveness and safety of the drug. Lack of efficacy and uncontrollable toxicity are the major causes of drug failures. Advances in drug distribution property prediction, particularly through in silico methods, have reduced screening time and costs. This study provides comprehensive knowledge on drug distribution, including influencing factors and artificial intelligence-based prediction models. The review also presents future challenges and research directions, aiming to facilitate innovative approaches in drug discovery.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Medicine, General & Internal
Yingwei Guo, Yingjian Yang, Fengqiu Cao, Wei Li, Mingming Wang, Yu Luo, Jia Guo, Asim Zaman, Xueqiang Zeng, Xiaoqiang Miu, Longyu Li, Weiyan Qiu, Yan Kang
Summary: This study evaluated the performance of clinical text information, radiomics features, and survival features for predicting functional outcomes of ischemic stroke patients. The results show that combining these features can improve the prediction accuracy, but further validation on larger and more varied datasets is necessary.
Review
Medicine, General & Internal
Sergio de Jesus Romero-Tapia, Jose Raul Becerril-Negrete, Jose A. Castro-Rodriguez, Blanca E. Del-Rio-Navarro
Summary: The clinical manifestations of asthma in children are diverse and associated with different molecular and cellular mechanisms. Early identification of high-risk patients and prediction models are essential. This review summarizes predictive factors, including lung function, allergic comorbidity, and relevant data from patient's medical history. It also highlights epigenetic factors such as DNA methylation, microRNA expression, and histone modification.
JOURNAL OF CLINICAL MEDICINE
(2023)
Review
Pharmacology & Pharmacy
Haoyang Liu, Zhiguang Fan, Jie Lin, Yuedong Yang, Ting Ran, Hongming Chen
Summary: Drug combination therapy is a common strategy for treating complex diseases. Efficient identification of appropriate drug combinations using computational methods is urgently needed due to the high cost of experimental screening. Recent studies have shown that deep learning algorithms have the flexibility to integrate multimodal data and achieve state-of-the-art performance, making deep-learning-based prediction of drug combinations an important tool in future drug discovery.
DRUG DISCOVERY TODAY
(2023)
Article
Oncology
Lishui Niu, Xianjing Chu, Xianghui Yang, Hongxiang Zhao, Liu Chen, Fuxing Deng, Zhan Liang, Di Jing, Rongrong Zhou
Summary: A multiomics model was developed to predict the risk of radiation pneumonitis (RP) in lung cancer patients, and its impact on survival was investigated. The results showed that the multiomics model accurately predicted the risk of RP, and RP patients had longer overall survival, especially those with mild RP.
JOURNAL OF CANCER RESEARCH AND CLINICAL ONCOLOGY
(2023)
Article
Biology
Quincy A. Hathaway, Naveena Yanamala, Matthew J. Budoff, Partho P. Sengupta, Irfan Zeb
Summary: The study followed 6814 participants for 16 years and found that the DeepSurv model significantly outperformed the COXPH model in ASCVD risk prediction, accurately predicting MAE and mortality. Results showed that DeepSurv was the only learning algorithm to significantly improve risk score criteria.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Genetics & Heredity
Yiran E. Liu, Sirle Saul, Aditya Manohar Rao, Makeda Lucretia Robinson, Olga Lucia Agudelo Rojas, Ana Maria Sanz, Michelle Verghese, Daniel Solis, Mamdouh Sibai, Chun Hong Huang, Malaya Kumar Sahoo, Rosa Margarita Gelvez, Nathalia Bueno, Maria Isabel Estupinan Cardenas, Luis Angel Villar Centeno, Elsa Marina Rojas Garrido, Fernando Rosso, Michele Donato, Benjamin A. Pinsky, Shirit Einav, Purvesh Khatri
Summary: The study integrated multiple datasets and developed an XGBoost model based on 8 genes, which accurately predicted the progression of severe dengue. The model performed well in predicting during the early febrile stage.
Article
Biology
Rakesh Chandra Joshi, Rashmi Mishra, Puneet Gandhi, Vinay Kumar Pathak, Radim Burget, Malay Kishore Dutta
Summary: The study proposed a novel two-stage ensemble of an ensemble-type machine learning-based predictive framework for glioma detection and its histograde classification. By considering different characteristics and applying machine learning approaches, the study achieved high accuracy and other statistical parameters.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Review
Biochemical Research Methods
Christoph A. Krettler, Gerhard G. Thallinger
Summary: Metabolomics and lipidomics play significant roles in personalized medicine, but data analysis remains a bottleneck. Recent research utilizing computational methods and simulating mass spectra has made valuable progress in annotating experimental data.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Clinical Neurology
Shraddha Mainali, Marin E. Darsie, Keaton S. Smetana
Summary: The application of machine learning in medicine has advanced rapidly over the past decade, with commercially available algorithms already being used in clinical settings for rapid diagnoses in stroke cases. The development of deep learning techniques has greatly improved the accuracy of stroke diagnosis and outcome prediction, although challenges still remain due to the multitude of factors influencing stroke prognosis. Despite the potential of machine learning, there are limitations such as small sample sizes in studies, emphasizing the need for more data collection and evaluation to enhance the efficacy of these tools in stroke management.
FRONTIERS IN NEUROLOGY
(2021)
Review
Biochemistry & Molecular Biology
Kira S. Makarova, Yuri I. Wolf, Eugene Koonin
Summary: Many viruses infecting bacteria and archaea produce proteins that inhibit the CRISPR-Cas system called anti-CRISPR proteins (Acr). These Acrs are highly specific for certain CRISPR variants, making their prediction and identification challenging. Discovering and characterizing Acrs is important not only for understanding the evolution of defense systems in prokaryotes, but also for the development of CRISPR-based biotechnological tools.
JOURNAL OF MOLECULAR BIOLOGY
(2023)
Article
Biology
Yunwei Zhang, Germaine Wong, Graham Mann, Samuel Muller, Jean Y. H. Yang
Summary: This article introduces the importance of survival analysis and proposes a new benchmarking design, SurvBenchmark, for evaluating the performance of various survival models on clinical and omics datasets. Through a systematic comparison of 320 comparisons, it demonstrates the variations of survival models in real-world applications and highlights the importance of using multiple performance metrics for evaluation.
Article
Biochemical Research Methods
Sehwan Moon, Hyunju Lee
Summary: This article introduces a multi-task attention learning algorithm, MOMA, for multi-omics data, which achieves high diagnostic performance and interpretability by capturing important biological processes. Experimental results demonstrate the superior performance of MOMA in various classification tasks, and its utility is verified through comparison experiments and biological analysis.
Article
Biochemistry & Molecular Biology
So-Young Yoon, Eunhong Jang, Naho Ko, Minseok Kim, Su Yoon Kim, Yeojin Moon, Jeong-Seok Nam, Sunjae Lee, Youngsoo Jun
Summary: The asymmetry in cytosolic pH between mother and daughter cells is believed to be responsible for cellular aging in budding yeast. Preferential accumulation of Pma1p in mother cells, which reduces the level of cytoplasmic protons, is thought to contribute to this pH increase. However, this study found that the accumulation of Pma1p in mother cells is not the key determinant of aging.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2022)
Article
Gastroenterology & Hepatology
Vishal C. Patel, Sunjae Lee, Mark J. W. McPhail, Kevin Da Silva, Susie Guilly, Ane Zamalloa, Elizabeth Witherden, Sidsel Stoy, Godhev Kumar Manakkat Vijay, Nicolas Pons, Nathalie Galleron, Xaiohong Huang, Selin Gencer, Muireann Coen, Thomas Henry Tranah, Julia Alexis Wendon, Kenneth D. Bruce, Emmanuelle Le Chatelier, Stanislav Dusko Ehrlich, Lindsey Ann Edwards, Saeed Shoaie, Debbie Lindsay Shawcross
Summary: Rifaximin-alpha can alleviate hepatic encephalopathy and reduce the likelihood of infection by reducing systemic inflammation and promoting gut barrier repair.
JOURNAL OF HEPATOLOGY
(2022)
Article
Multidisciplinary Sciences
Jihee Soh, Sejin Park, Hyunju Lee
Summary: We developed a novel method called HIDTI for predicting interactions between drugs and disease-causing proteins. By using a residual network to extract features from heterogeneous information, our method accurately predicts the targets of new drugs.
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Hyejin Cho, Baeksoo Kim, Wonjun Choi, Doheon Lee, Hyunju Lee
Summary: This study presents a plant-phenotype relationship corpus that supports the development of natural language processing. The corpus includes a large amount of information related to plants and phenotypes, demonstrating significant performance of NLP in the test set.
Article
Oncology
Byong-Sop Song, Ji Sun Moon, Jingwen Tian, Ho Yeop Lee, Byeong Chang Sim, Seok-Hwan Kim, Seul Gi Kang, Jung Tae Kim, Ha Thi Nga, Rui Benfeitas, Yeongmin Kim, Sanghee Park, Robert R. Wolfe, Hyuk Soo Eun, Minho Shong, Sunjae Lee, Il-Young Kim, Hyon-Seung Yi
Summary: Mitochondrial ribosomal protein dysfunction is associated with hepatocellular carcinoma progression and leads to an immunometabolic microenvironment favorable for cancer progression. Impaired mitoribosomal function promotes glucose partitioning toward glycolytic flux, lactate synthesis, and T cell exhaustion. This study provides insights into the critical role of mitoribosomes in regulating the immunometabolic characteristics of liver cancer.
JOURNAL FOR IMMUNOTHERAPY OF CANCER
(2022)
Article
Oncology
Namjeong Choi, Ha Na Jang, Jagyeong Oh, Jiyeon Ha, Hyungbin Park, Xuexiu Zheng, Sunjae Lee, Haihong Shen
Summary: This study investigated the alternative splicing of the Fas gene and identified SRSF6 as a key regulator in this process. The results also showed that the correlation between SRSF6 and Fas expression differs between normal tissues and tumors. This research reveals a novel regulatory mechanism of Fas alternative splicing.
Article
Multidisciplinary Sciences
Ceri Proffitt, Gholamreza Bidkhori, Sunjae Lee, Abdellah Tebani, Adil Mardinoglu, Mathias Uhlen, David L. Moyes, Saeed Shoaie
Summary: The human gut microbiome is associated with metabolic disorders such as obesity, type 2 diabetes, and atherosclerosis. The study investigated the role of gut bacteria in metabolic diseases using metagenomics data and metabolic modeling. The modeling predicted changes in glutamate consumption and the production of ammonia, arginine, and proline in gut bacteria common across the disorders. The study also found that tartrate dehydrogenase is enriched in the disorders and an increased tartrate metabolism is associated with certain metabolites in healthy obese individuals.
Article
Mathematical & Computational Biology
Bouchra Ezzamouri, Dorines Rosario, Gholamreza Bidkhori, Sunjae Lee, Mathias Uhlen, Saeed Shoaie
Summary: The human gut microbiome plays a significant role in metabolic disorders, such as type 2 diabetes mellitus. This study investigates the mechanistic role of the gut microbiome in response to metformin treatment, using metagenomics data and genome-scale metabolic modeling. The results demonstrate the commensal and competing behavior of key bacterial species in response to metformin and highlight the impact of different nutritional environments.
NPJ SYSTEMS BIOLOGY AND APPLICATIONS
(2023)
Article
Endocrinology & Metabolism
Jibeom Lee, Jiseung Kang, Yumin Kim, Sunjae Lee, Chang-Myung Oh, Tae Kim
Summary: This study investigated the effects of sleep deprivation (SD) and high-fat diet (HFD)-induced obesity on gut microbiota and host responses. The results showed that HFD significantly altered the gut microbiota, while SD had a major impact on the gut transcriptome. When combined, SD and HFD severely disrupted the brain's inflammatory system. In addition, inosine-5' phosphate may be a key metabolite mediating microbiota-gut-brain interactions.
FRONTIERS IN ENDOCRINOLOGY
(2023)
Review
Chemistry, Medicinal
Xinjie Song, Nikhil Kirtipal, Sunjae Lee, Petr Maly, Shiv Bharadwaj
Summary: Caffeine acts as a nonselective adenosine receptor antagonist and has both beneficial effects and paradoxical effects on human health. This article provides an overview of caffeine's validated targets and its impact on organ-specific physiology and pathophysiology. Further studies are needed to explore caffeine-induced changes in specific conditions for therapeutic applications.
PHYTOTHERAPY RESEARCH
(2023)
Meeting Abstract
Endocrinology & Metabolism
J. Son, J. Ahn, O. -K. Hong, S. Lee, S. Chung
Article
Biochemical Research Methods
Sehwan Moon, Jeongyoung Hwang, Hyunju Lee
Summary: Integration of multi-omics data using the proposed supervised deep generalized canonical correlation analysis (SDGCCA) method improves phenotypic classification and biomarker identification. By considering complex/nonlinear cross-data correlations between multiple modalities, SDGCCA outperforms other methods in predicting Alzheimer's disease (AD) and discriminating early- and late-stage cancers. Additionally, SDGCCA enables feature selection and identifies important multi-omics biomarkers associated with AD.
JOURNAL OF COMPUTATIONAL BIOLOGY
(2022)