Article
Multidisciplinary Sciences
Zhi Zhao, Shixiong Wang, Manuela Zucknick, Tero Aittokallio
Summary: Researchers developed a mix-lasso model that accurately predicts drug response and identifies tissue-specific predictive features, addressing the limitations of current statistical models in leveraging various cancer tissues and multi-omics profiles.
Article
Biology
Hua Chai, Xiang Zhou, Zhongyue Zhang, Jiahua Rao, Huiying Zhao, Yuedong Yang
Summary: The study employed denoising Autoencoder to integrate multi-omics data, improving cancer prognosis prediction accuracy. Results showed a 6.5% average increase in C-index values compared to previous methods across 15 cancers, and successfully differentiated high-risk and low-risk patients in the breast cancer case study.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Biochemical Research Methods
Xingze Wang, Guoxian Yu, Jun Wang, Azlan Mohd Zain, Wei Guo
Summary: Diagnosing lung cancer subtypes accurately is crucial for precise treatment. This study introduces an interpretable and flexible solution called LungDWM, which utilizes weakly paired multiomics data to diagnose lung cancer subtypes. By extracting important diagnostic features, imputing missing data, and fusing information from different omics, LungDWM outperforms other competitive methods in terms of performance, authenticity, and interpretability.
Article
Biochemical Research Methods
Gang Wen, Limin Li
Summary: In this work, a novel multi-omics deep survival prediction approach named FGCNSurv is proposed, which utilizes a dually fused graph convolutional network (GCN). The FGCNSurv method demonstrates superior performance in extracting complementary information from multi-omics data, and outperforms existing survival prediction methods on real-world datasets.
Review
Biochemical Research Methods
Susana Vinga
Summary: The development of new molecular and cell technologies is generating a large amount of data, leading to a growth in omics databases and presenting challenges for statistical learning and computational biology in health applications. Regularized optimization methods have emerged as a promising strategy to address the high dimensionality of these data and improve accuracy in building models for biological observations.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Genetics & Heredity
Zongliang Hu, Yan Zhou, Tiejun Tong
Summary: A robust variable selection algorithm based on logistic regression was developed for meta-analyzing high-dimensional datasets, using a combination of least trimmed squared estimates and hierarchical bi-level variable selection technique, to achieve more reliable results.
FRONTIERS IN GENETICS
(2021)
Article
Automation & Control Systems
Daniel R. Kowal
Summary: Subset selection is a valuable tool for interpretability, scientific discovery, and data compression. We propose a Bayesian approach to address the challenges in classical subset selection, and introduce a strategy that focuses on finding near-optimal subsets rather than a single best subset. We apply Bayesian decision analysis to derive the optimal linear coefficients for any subset of variables, and our approach outperforms competing methods in prediction, interval estimation, and variable selection. By analyzing a large education dataset, we gain unique insights into the factors that predict educational outcomes and identify over 200 distinct subsets of variables that offer near-optimal predictive accuracy.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Medicine, Research & Experimental
Wei Chong, Xingyu Zhu, Huicheng Ren, Chunshui Ye, Kang Xu, Zhe Wang, Shengtao Jia, Liang Shang, Leping Li, Hao Chen
Summary: This study elucidates the molecular heterogeneity of KRAS-mutant CRC and reveals new biological subtypes and therapeutic possibilities for these tumors.
Article
Biochemical Research Methods
Yongqing Zhang, Shuwen Xiong, Zixuan Wang, Yuhang Liu, Hong Luo, Beichen Li, Quan Zou
Summary: Cancer prognosis prediction and analysis can help patients understand expected life and help clinicians provide correct therapeutic guidance. Graph neural networks have become mainstream in cancer prognosis prediction and analysis, but their accuracy is limited by the number of neighboring genes in biological networks. To solve this problem, a local augmented graph convolutional network named LAGProg is proposed in this paper.
Article
Biotechnology & Applied Microbiology
Vidhi Malik, Yogesh Kalakoti, Durai Sundar
Summary: The study proposed a late multi-omics integrative framework that effectively quantifies survival and drug response for breast cancer patients, comparing the predictive abilities of different omics data types. Using features selected by NCA, a neural network framework successfully developed models for predicting survival and drug response.
Article
Multidisciplinary Sciences
Sven E. Ojavee, Athanasios Kousathanas, Daniel Trejo Banos, Etienne J. Orliac, Marion Patxot, Kristi Laell, Reedik Maegi, Krista Fischer, Zoltan Kutalik, Matthew R. Robinson
Summary: This study introduces a Bayesian approach for probabilistic inference of the genetic architecture of age-at-onset phenotypes, demonstrating its benefits in simulation work and data from the UK Biobank.
NATURE COMMUNICATIONS
(2021)
Article
Multidisciplinary Sciences
Conghao Wang, Wu Lue, Rama Kaalia, Parvin Kumar, Jagath C. Rajapakse
Summary: This paper evaluates two network-based approaches for integrating multi-omics data in the prediction of clinical outcomes for neuroblastoma. The results show that network-level fusion outperforms feature-level fusion, and network-based methods are effective in handling heterogeneity and high dimensionality in multi-omics data integration.
SCIENTIFIC REPORTS
(2022)
Article
Multidisciplinary Sciences
Zhaoxiang Cai, Rebecca C. Poulos, Jia Liu, Qing Zhong
Summary: This article reviews the application of machine learning in multi-omics data analysis, including both general-purpose and task-specific methods. By benchmarking the performance of five machine learning approaches using data from the Cancer Cell Line Encyclopedia, recommendations are provided for method selection in specific applications. The importance of this research lies in promoting the development of novel machine learning methodologies, which will be critical for drug discovery, clinical trial design, and personalized treatments.
Article
Agronomy
Dominic Knoch, Christian R. Werner, Rhonda C. Meyer, David Riewe, Amine Abbadi, Sophie Luecke, Rod J. Snowdon, Thomas Altmann
Summary: Using transcriptome data and reproducing kernel Hilbert space regression based on Gaussian kernels can enhance hybrid prediction accuracies for complex agronomic traits in canola, surpassing the predictive abilities of genetic markers alone. The study demonstrates that transcripts contain valuable information beyond genomic data, with potential implications for future canola breeding programs.
THEORETICAL AND APPLIED GENETICS
(2021)
Article
Mathematical & Computational Biology
Wei Li, Binchun Liu, Weiqian Wang, Can Sun, Jianpeng Che, Xuelian Yuan, Chunbo Zhai
Summary: This study used the random forest algorithm for lung cancer stage prediction and found that the prediction accuracy of microbial combinatorial transcriptome fusion analysis was the highest. This can assist doctors in accurately diagnosing lung cancer stage in clinics.
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
(2022)
Article
Genetics & Heredity
Xujun Wang, Zhengtao Zhang, Weny Qiu, Shiyi Liu, Cong Liu, Georgi Z. Genchev, Lijian Hui, Hongyu Zhao, Hui Lu
Summary: RePhine is a regression-based pharmacogenomic and ChIP-seq data integration method that infers the impact of transcriptional regulators on drug response. Evaluation on simulation and pharmacogenomic data showed improved performance of RePhine in identifying drug response-related transcriptional regulators.
GENOMICS PROTEOMICS & BIOINFORMATICS
(2021)
Article
Health Care Sciences & Services
Mengting Ji, Georgi Z. Genchev, Hengye Huang, Ting Xu, Hui Lu, Guangjun Yu
Summary: This study aimed to develop and validate a measurement instrument and test the interrelationships of evaluation variables for an artificial intelligence-enabled clinical decision support system evaluation framework. The results showed that user acceptance is the central dimension of artificial intelligence-enabled clinical decision support system success, directly influenced by perceived ease of use, information quality, service quality, and perceived benefit, and indirectly influenced through system quality and information quality.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Cell Biology
Yu Wang, Yiyi Liang, Haiyan Xu, Xiao Zhang, Tiebo Mao, Jiujie Cui, Jiayu Yao, Yongchao Wang, Feng Jiao, Xiuying Xiao, Jiong Hu, Qing Xia, Xiaofei Zhang, Xujun Wang, Yongwei Sun, Deliang Fu, Lei Shen, Xiaojiang Xu, Jing Xue, Liwei Wang
Summary: The study investigated the heterogeneity among PDAC patients in terms of CAFs, ductal cancer cells, and immune cells, revealing differences between dense-type and loose-type PDAC. A novel subtype of CAFs, meCAFs, was discovered in loose-type PDAC, playing a critical role in disease progression and response to immunotherapy. This finding highlights the importance of considering intertumoral heterogeneity in PDAC treatment strategies.
Article
Genetics & Heredity
Jianlei Gu, Jiawei Dai, Hui Lu, Hongyu Zhao
Summary: Comprehensive characterization of spatial and temporal gene expression patterns in humans is critical for understanding human diseases. This study proposed a data-driven framework to derive a list of Ubiquitously Expressed Genes (UEGs) and their global expression patterns, providing a valuable resource for further characterizing the human transcriptome.
GENOMICS PROTEOMICS & BIOINFORMATICS
(2023)
Article
Education, Scientific Disciplines
Erin M. White, Andrew C. Esposito, Vadim Kurbatov, Xujun Wang, Michael G. Caty, Maxwell Laurans, Peter S. Yoo
Summary: This study examines the variability of surgical attending experience and perspectives on informed consent, and how it impacts resident education. The results show that surgeon's demographics, personal experiences, and specialty significantly influence their teaching styles and the educational experience residents receive regarding informed consent.
JOURNAL OF SURGICAL EDUCATION
(2022)
Article
Psychiatry
Yuanyuan Gui, Xiaocheng Zhou, Zixin Wang, Yiliang Zhang, Zhaobin Wang, Geyu Zhou, Yize Zhao, Manhua Liu, Hui Lu, Hongyu Zhao
Summary: This study examined the associations between the predicted polygenic risk scores of six psychiatric disorders and cognitive functions, behavior, and brain imaging traits using publicly available genome-wide association studies and individual-level data from the Adolescent Brain Cognitive Development study and the UK Biobank study. Significant heterogeneity was found in genetic associations between cognitive traits and psychiatric disorders, as well as behavior and brain imaging, between sexes.
TRANSLATIONAL PSYCHIATRY
(2022)
Article
Biology
Yongyong Ren, Yan Kong, Xiaocheng Zhou, Georgi Z. Genchev, Chao Zhou, Hongyu Zhao, Hui Lu
Summary: The quality control of genetic variants from whole-genome sequencing data is important in clinical diagnosis and human genetics research. The current filtering methods have limitations, but FVC, an adaptive method, performs better in filtering out false variants.
COMMUNICATIONS BIOLOGY
(2022)
Article
Psychiatry
Zhaobin Wang, Xiaocheng Zhou, Yuanyuan Gui, Manhua Liu, Hui Lu
Summary: Attention deficit hyperactivity disorder (ADHD) is a common psychiatric disorder in school-aged children. This study proposes an automated ADHD classification framework by combining multiple measures of resting-state functional magnetic resonance imaging (rsfMRI) in the adolescent brain.
TRANSLATIONAL PSYCHIATRY
(2023)
Article
Oncology
Tong Han, Xujun Wang, Sailing Shi, Wubing Zhang, Jue Wang, Qiu Wu, Ziyi Li, Jingxin Fu, Rongbin Zheng, Jiamin Zhang, Qin Tang, Peng Zhang, Chenfei Wang
Summary: This study performed IFNy sensitivity screens in over 40 cancer cell lines and discovered that activation of double-strand break repair genes could lead to resistance to IFNy in cancer cells. Suppression of single-strand break repair genes increased the reliance on DSB repair genes after IFNy treatment. Inhibition of the DSB repair pathway exhibited a synergistic effect with IFNy treatment both in vitro and in vivo. The relationship between the activation of DSB repair genes and IFNy resistance was further confirmed in clinical tumor profiles. This study provides comprehensive resources and evidence to elucidate a mechanism of IFNy resistance in cancer and has the potential to inform combination therapies to overcome immunotherapy resistance.
CANCER IMMUNOLOGY RESEARCH
(2023)
Article
Health Care Sciences & Services
Shuya Cui, Qingmin Lin, Yuanyuan Gui, Yunting Zhang, Hui Lu, Hongyu Zhao, Xiaolei Wang, Xinyue Li, Fan Jiang
Summary: In this study, a new metric called circadian activity rhythm energy (CARE) is introduced to better measure circadian amplitude. CARE is found to be significantly correlated with melatonin amplitude and associated with cognitive functions. The study also identifies a genetic basis for CARE and demonstrates its causal effect on cognitive functions.
NPJ DIGITAL MEDICINE
(2023)
Article
Computer Science, Artificial Intelligence
Shu Wang, Xiaocheng Zhou, Yan Kong, Hui Lu
Summary: Spatially resolved transcriptomics (SRT) is a vital technique for measuring gene expression while preserving spatial information. In this study, researchers propose a deep learning-based method to enhance spot resolution, achieving higher resolution SRT data. The method outperforms traditional superresolution techniques and offers a substantial advancement in SRT by enabling higher-resolution gene expression data generation.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Fan Yang, Wenchuan Wang, Fang Wang, Yuan Fang, Duyu Tang, Junzhou Huang, Hui Lu, Jianhua Yao
Summary: Annotating cell types based on single-cell RNA-seq data is crucial for studying disease progress and tumor microenvironments. Existing annotation methods often face challenges such as a lack of curated marker gene lists, difficulties in handling batch effects, and inability to leverage gene-gene interaction information, leading to limited generalization and robustness. In this study, we developed a pretrained deep neural network model called scBERT, which overcomes these challenges by training on large amounts of unlabeled scRNA-seq data and achieving a comprehensive understanding of gene-gene interactions. scBERT demonstrates superior performance in cell type annotation, novel cell type discovery, robustness to batch effects, and model interpretability.
NATURE MACHINE INTELLIGENCE
(2022)
Article
Mathematical & Computational Biology
Chenfang Zhang, Georgi Z. Genchev, Dennis Bergau, Hui Lu
INTERDISCIPLINARY SCIENCES-COMPUTATIONAL LIFE SCIENCES
(2020)
Article
Cardiac & Cardiovascular Systems
Liangdong Sun, Jie Dai, Xujun Wang, Gening Jiang, Diego Gonzalez-Rivas, Jiong Song, Peng Zhang
INTERACTIVE CARDIOVASCULAR AND THORACIC SURGERY
(2020)
Article
Biochemical Research Methods
K. Ramki, G. Thiruppathi, Selva Kumar Ramasamy, P. Sundararaj, P. Sakthivel
Summary: A chromone-based ratiometric fluorescent probe L2 was developed for the selective detection of Hg(II) in a semiaqueous solution. The probe exhibited enhanced fluorescence in its aggregated state and even higher fluorescence when chelated with Hg(II). The probe demonstrated high sensitivity and specificity for Hg(II) detection and was successfully applied for imaging Hg(II) in a living model.
Article
Biochemical Research Methods
Qun Zhang, Rui Yang, Gang Liu, Shiyan Jiang, Jiarui Wang, Juqiang Lin, Tingyin Wang, Jing Wang, Zufang Huang
Summary: This research aims to develop a cost-effective and portable method for measuring creatinine levels using the enhanced Tyndall effect phenomenon. The method offers a promising solution for monitoring renal healthcare in resource-limited settings.