Article
Biochemistry & Molecular Biology
Yingnan Song, Zhe Yin, Chuan Zhang, Shengju Hao, Haibo Li, Shifan Wang, Xiangchun Yang, Qiong Li, Danyan Zhuang, Xinyuan Zhang, Zongfu Cao, Xu Ma
Summary: This article describes a PKU screening model using a random forest classifier, which demonstrates excellent performance in validation datasets and two Chinese testing populations, contributing to early diagnosis and prevention of PKU.
FRONTIERS IN MOLECULAR BIOSCIENCES
(2022)
Article
Mathematics, Applied
Serpil Yalcin Kuzu
Summary: Data used in particle physics analyses often have imbalanced nature, making it difficult to identify rare events of interest from the background. This study explores the use of supervised machine learning approaches, specifically classification algorithms, to interpret skewed particle datasets. The application of a multiclass classification approach based on random forest classifier (RFC) showed promising results in the analysis of the ground state and excited states of bottomonium.
JOURNAL OF SCIENTIFIC COMPUTING
(2023)
Article
Biochemical Research Methods
Tao Zhou, Libin Chen, Jing Guo, Mengmeng Zhang, Yanrui Zhang, Shanbo Cao, Feng Lou, Haijun Wang
Summary: MSIFinder, a Python package using random forest classifier (RFC)-based genome sequencing, is a robust and effective tool for MSI classification, achieving high sensitivity and specificity in detecting MSI, regardless of sequencing depth and panel size influences.
BMC BIOINFORMATICS
(2021)
Article
Computer Science, Information Systems
Darren Yates, Md Zahidul Islam
Summary: The FastForest algorithm, with its three optimizing components, achieves faster processing speed on hardware-constrained devices while maintaining high accuracy, suitable for both PC and smartphone platforms. Empirical testing shows excellent performance against other ensemble classifiers, surpassing them in various tests.
INFORMATION SCIENCES
(2021)
Article
Medicine, General & Internal
Jian Yu, Xiaoyan Xie, Yun Zhang, Feng Jiang, Chuyan Wu
Summary: Obesity is a global health concern that increases the risk of chronic diseases and reduces life expectancy and quality of life. Traditional diagnosis methods for obesity have flaws, necessitating the design of new diagnostic models. Recent advancements in gene sequencing technology have led to the discovery of more obesity-related markers. Using gene expression profiles, 12 important genes associated with obesity were identified. An artificial neural network was also used to develop an effective obesity diagnosis model.
FRONTIERS IN MEDICINE
(2022)
Article
Forestry
Yihua Jin, Jingrong Zhu, Guishan Cui, Zhenhao Yin, Weihong Zhu, Dong Kun Lee
Summary: This study aims to characterize forest cover transitions and identify degraded or at-risk areas in North Korea. Using phenological information and random forest classifiers, a deforestation classification was performed. The analysis of deforestation dynamics from 1990 to 2020 revealed severe degradation and fragmentation of forests in North Korea.
Article
Mathematics, Applied
Vijay K. Yadav, Nilam
Summary: Precision in the measurement of glucose levels in the artificial pancreas is crucial. Machine learning techniques, such as decision tree, random forest, support vector machine, and K-nearest neighbor, are proposed for predicting and classifying diabetes mellitus. The KNN model displays the highest accuracy in comparison to others.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2023)
Article
Oncology
Ariana Rostami, Scott Bratman, Kathy Han
Summary: Liquid biopsy approaches for detecting viral DNA play a crucial role in diagnosing virally-associated cancers. The CaptHPV method based on next-generation sequencing is discussed for its potential in detecting plasma HPV DNA in HPV-associated cancers and its potential clinical utility.
CLINICAL CANCER RESEARCH
(2021)
Article
Environmental Sciences
Luciana Nieto, Rasmus Houborg, Ariel Zajdband, Arin Jumpasut, P. V. Vara Prasad, Brad J. S. C. Olson, Ignacio A. Ciampitti
Summary: It is crucial for farmers, policymakers, and government agencies to accurately define agricultural crop phenology and its spatial-temporal variability. This study proposes using high-cadence earth observations and robust classifiers to improve the accuracy of crop phenology classification. The findings suggest that high temporal resolution data significantly enhances crop classification metrics compared to lower temporal resolution data. Additionally, the research emphasizes the criticality of high temporal resolution earth observation data for agriculture decision making.
Article
Environmental Sciences
Emma C. Hall, Mark J. Lara
Summary: Uncrewed aerial systems (UASs) have been proven to be powerful tools for ecological observations, especially in measuring plant physiological and phenological traits. However, the high cost of drone-borne sensors limits their widespread use. This study evaluates the tradeoffs between off-the-shelf and sophisticated sensors for mapping plant species and functional types in a diverse grassland. The results show that off-the-shelf multispectral sensors can achieve comparable mapping accuracies by integrating phenometrics into machine learning image classifiers.
Article
Cell Biology
Alireza Rouzitalab, Chadwick B. Boulay, Jeongwon Park, Julio C. Martinez-Trujillo, Adam J. Sachs
Summary: The neuronal ensembles in the lateral prefrontal cortex of non-human primates can dynamically encode and store arbitrary stimulus-response associations. These ensembles rapidly learn new associations and can retrieve multiple previously learned associations from a neuronal subspace. Additionally, knowledge of old associations facilitates the learning of new, similar associations.
Article
Environmental Sciences
Panpan Wei, Weiwei Zhu, Yifan Zhao, Peng Fang, Xiwang Zhang, Nana Yan, Hao Zhao
Summary: This study used Kenya as the study area and employed machine learning algorithms and remote sensing data to accurately and quickly map grasslands, providing important support for the stable development of the local animal husbandry economy. The research identified the optimal feature combination and classification method, laying the foundation for future land cover classification.
Article
Environmental Sciences
Xiaochun Zhai, Rui Xu, Zhixiong Wang, Zhaojun Zheng, Yixuan Shou, Shengrong Tian, Lin Tian, Xiuqing Hu, Lin Chen, Na Xu
Summary: The Ku-band scatterometer called CSCAT onboard the Chinese-French Oceanography Satellite (CFOSAT) is the first spaceborne rotating fan-beam scatterometer (RFSCAT). A new algorithm using a random forest classifier is presented to classify Arctic sea ice types based on CSCAT measurement data. The algorithm extracts innovative feature parameters from orbital measurement for the first time and achieves high overall accuracy and precision for water, first-year ice (FYI), and multi-year ice (MYI). The algorithm is validated and compared with other datasets, showing good spatial homogeneity and improved detection of MYI mobility in the East Greenland region.
Article
Biotechnology & Applied Microbiology
Eric M. Davis, Yu Sun, Yanling Liu, Pandurang Kolekar, Ying Shao, Karol Szlachta, Heather L. Mulder, Dongren Ren, Stephen V. Rice, Zhaoming Wang, Joy Nakitandwe, Alexander M. Gout, Bridget Shaner, Salina Hall, Leslie L. Robison, Stanley Pounds, Jeffery M. Klco, John Easton, Xiaotu Ma
Summary: The study proposed a new computational method, SequencErr, to measure errors in sequencing instruments, revealing the sequencer error rate to be around 10 per million. The method demonstrated a 10-fold lower error rate compared to popular error correction methods and can provide novel insights into DNA sequencing errors.
Article
Virology
Jaime Leonardo Moreno-Gallego, Alejandro Reyes
Summary: Viruses play a significant role in microbial ecosystems, and research on viral taxonomy has made progress through the identification of potential markers for studying viral genetic information. By studying and clustering common features of viral genes, a set of signature ViPhOGs has been established.