Filters








447 Hits in 8.4 sec

Component-wise gradient boosting and false discovery control in survival analysis with high-dimensional covariates

Kevin He, Yanming Li, Ji Zhu, Hongliang Liu, Jeffrey E. Lee, Christopher I. Amos, Terry Hyslop, Jiashun Jin, Huazhen Lin, Qinyi Wei, Yi Li
2015 Bioinformatics  
Specifically, we modified the component-wise gradient boosting to improve the computational feasibility and introduced random permutation in stability selection for controlling false discoveries.  ...  Results: We have proposed a high-dimensional variable selection method by incorporating stability selection to control false discovery.  ...  Component-wise gradient boosting and false discovery control In contrast, Algorithm 1 is based on gradient with respect to b and it avoids the calculation of working response.  ... 
doi:10.1093/bioinformatics/btv517 pmid:26382192 pmcid:PMC4757968 fatcat:npuzbntxvzg3rmc6432qk4lyhe

False Discovery Rate Control in Cancer Biomarker Selection Using Knockoffs

Arlina Shen, Han Fu, Kevin He, Hui Jiang
2019 Cancers  
Our method is general and flexible, accommodating arbitrary covariate distributions, linear and nonlinear associations, and survival models.  ...  To ensure that most of the discoveries are true, we employ a knockoff procedure to control false discoveries.  ...  Therefore they will be selected via univariate analysis, leading to high false discoveries.  ... 
doi:10.3390/cancers11060744 pmid:31146393 pmcid:PMC6628039 fatcat:3t4cl5qrsnccxhyio6u4mjmqiq

EM and component-wise boosting for Hidden Markov Models: a machine-learning approach to capture-recapture [article]

Robert William Rankin
2016 biorxiv/medrxiv   pre-print
, and spatial components.  ...  survival and abundance, yet yields qualitatively similar estimates.  ...  In other words, they incur False Discoveries.  ... 
doi:10.1101/052266 fatcat:sm5uhncc5bcdfkan3gcph25b2q

Novel image markers for non-small cell lung cancer classification and survival prediction

Hongyuan Wang, Fuyong Xing, Hai Su, Arnold Stromberg, Lin Yang
2014 BMC Bioinformatics  
Finally, a Cox proportional hazards model is fitted by component-wise likelihood based boosting.  ...  The entire biomedical imaging informatics framework consists of cell detection, segmentation, classification, discovery of image markers, and survival analysis.  ...  The project is also partially supported by the National Center for Research Resources and the National  ... 
doi:10.1186/1471-2105-15-310 pmid:25240495 pmcid:PMC4287550 fatcat:hp4vyhj7kjc7lkuenjjctf4xaa

Controlling false discoveries in high-dimensional situations: boosting with stability selection

Benjamin Hofner, Luigi Boccuto, Markus Göker
2015 BMC Bioinformatics  
Results: Stability selection with boosting was able to detect influential predictors in high-dimensional settings while controlling the given error bound in various simulation scenarios.  ...  It proved to work well in high-dimensional settings with more predictors than observations for both, linear and additive models.  ...  Schwartz from the Greenwood Genetic Center for their help with the analysis and with the interpretation of the results, as well as Michael Drey who conducted an early version of the presented simulation  ... 
doi:10.1186/s12859-015-0575-3 pmid:25943565 pmcid:PMC4464883 fatcat:ovcrv6uipvevnn4ycpds7wgslm

Changes in Social Position Predict Survival in Bottlenose Dolphins [article]

Robert William Rankin, Vivienne Jilla Foroughirad, Ewa Beata Krzyszczyk, Céline H Frère, Janet Mann
2022 bioRxiv   pre-print
We used two inferential frameworks to provide complimentary evidence for or against hypotheses; namely, a gradient-boosting predictivist approach with relative importance measures, and an inclusion probabilities  ...  This is critical to understand given the high variation in types of social structures and strategies within populations of social mammals, both across time and among individuals.  ...  Based on cross-validation performance, we decided to use component-wise gradient boosting.  ... 
doi:10.1101/2022.08.25.505273 fatcat:r5dlu3rj4veuvf5zr2akdtswca

Boosting Functional Regression Models with FDboost [article]

Sarah Brockhaus and David Rügamer and Sonja Greven
2018 arXiv   pre-print
Furthermore, boosting can be used in high-dimensional data settings with more covariates than observations.  ...  In addition to mean regression, quantile regression models as well as generalized additive models for location scale and shape can be fitted with FDboost.  ...  Component-wise gradient boosting allows to fit models in high-dimensional data situations and performs data-driven variable selection.  ... 
arXiv:1705.10662v3 fatcat:oureu7chw5gkvaexzgnr2p4uvy

Review of statistical methods for survival analysis using genomic data

Seungyeoun Lee, Heeju Lim
2019 Genomics & Informatics  
model with high-dimensional genomic data.  ...  We review traditional survival methods and regularization methods, with various penalty functions, for the analysis of high-dimensional genomics, and describe machine learning techniques that have been  ...  For coping with the problem of analyzing high-dimensional data, component-wise boosting has also been adapted to survival analysis.  ... 
doi:10.5808/gi.2019.17.4.e41 pmid:31896241 pmcid:PMC6944043 fatcat:dw7rubh7v5a3hcgptsyqnydk6a

Machine learning in drug discovery and development part 1 – a primer

Alan Talevi, Juan Francisco Morales, Gregory Hather, Jagdeep Podichetty, Sarah Kim, Peter C Bloomingdale, Samuel Kim, Jackson Burton, Joshua D Brown, Almut G Winterstein, Stephan Schmidt, J Kael White (+1 others)
2020 CPT: Pharmacometrics & Systems Pharmacology  
Here, we present a primer on the ML algorithms most commonly used in drug discovery and development.  ...  A companion article will summarize applications of ML in drug discovery, drug development, and postapproval phase.  ...  Dimensionality reduction methods may be applied as a first step in the analysis. 52 Principal component analysis is among the most widely used dimensionality reduction techniques.  ... 
doi:10.1002/psp4.12491 pmid:31905263 fatcat:5opymt6te5dstijlprl42jf6gy

Spatial parcellations, spectral filtering, and connectivity measures in fMRI: Optimizing for discrimination

Roser Sala-Llonch, Stephen M. Smith, Mark Woolrich, Eugene P. Duff
2018 Human Brain Mapping  
Partial correlation, with appropriate regularization, outperformed correlation. Amplitude and covariance generally discriminated less well, although gave good results with high-dimensionality ICA.  ...  We tested a range of anatomical and functional parcellations, including the AAL atlas, parcellations derived from the Human Connectome Project and Independent Component Analysis (ICA) of many dimensionalities  ...  To control false positives, we use False Discovery Rate (FDR) with q = 0.05, and test for robustness to variations of parcellation schemes using the AAL as a baseline (i.e., which parcellations give significantly  ... 
doi:10.1002/hbm.24381 pmid:30259597 pmcid:PMC6492132 fatcat:egknkah2hvg6vdfiuyz7nxrsla

Pathway analysis using random forests with bivariate node-split for survival outcomes

Herbert Pang, Debayan Datta, Hongyu Zhao
2009 Computer applications in the biosciences : CABIOS  
Thus, pathway-based survival analysis using machine learning tools represents a promising approach in dissecting pathways and for generating new biological hypothesis from microarray studies.  ...  Results: In this article, we describe a pathway-based method using random forests to correlate gene expression data with survival outcomes and introduce a novel bivariate node-splitting random survival  ...  Comparisons with other machine learning methods We compared the random forests approach and several machine learning tools for survival data in simulations, including gradient boosting with component-wise  ... 
doi:10.1093/bioinformatics/btp640 pmid:19933158 pmcid:PMC2804301 fatcat:wekexdiby5avdeinex5g3zvbfq

Computational Methods for the Discovery of Metabolic Markers of Complex Traits

Michael Lee, Ting Hu
2019 Metabolites  
budding with high potential for utility in metabolomics research.  ...  In this review, the workflow of metabolic marker discovery is outlined from metabolite extraction to model interpretation and validation.  ...  the classification. [25] Gradient boosting machine (GBM) Build an ensemble of decision trees in a step-wise fashion using boosting and gradient descent algorithms. [62] Artificial neural network (ANN)  ... 
doi:10.3390/metabo9040066 pmid:30987289 pmcid:PMC6523328 fatcat:iluqijb2pvgytkdc2m4gjmdtia

Multi-site study of additive genetic effects on fractional anisotropy of cerebral white matter: Comparing meta and megaanalytical approaches for data pooling

Peter Kochunov, Neda Jahanshad, Emma Sprooten, Thomas E. Nichols, René C. Mandl, Laura Almasy, Tom Booth, Rachel M. Brouwer, Joanne E. Curran, Greig I. de Zubicaray, Rali Dimitrova, Ravi Duggirala (+29 others)
2014 NeuroImage  
We evaluated this protocol in five family-based cohorts providing data from a total of 2248 children and adults (ages: 9-85) collected with various imaging protocols.  ...  Combining datasets across independent studies can boost statistical power by increasing the numbers of observations and can achieve more accurate estimates of effect sizes.  ...  Acknowledgments This study was supported by R01 EB015611 to PK, R01 HD050735 to PT, MH0708143 and MH083824 grants to DCG and by MH078111 and MH59490 to JB.  ... 
doi:10.1016/j.neuroimage.2014.03.033 pmid:24657781 pmcid:PMC4043878 fatcat:mxe7lkp2hfdyrdmcgzzhdbh3zm

Flexible variable selection in the presence of missing data [article]

B. D. Williamson, Y. Huang
2022 arXiv   pre-print
In cases where this model is misspecified, the selected variables may not all be truly scientifically relevant and can result in panels with suboptimal classification performance.  ...  Through simulations, we show that our proposals have good operating characteristics and result in panels with higher classification performance compared to several existing penalized regression approaches  ...  The opinions expressed in this article are those of the authors and do not necessarily represent the official views of the NIH.  ... 
arXiv:2202.12989v2 fatcat:gl63axewjnbvngfau4annfq7ji

Improving prognostic performance in resectable pancreatic ductal adenocarcinoma using radiomics and deep learning features fusion in CT images

Yucheng Zhang, Edrise M. Lobo-Mueller, Paul Karanicolas, Steven Gallinger, Masoom A. Haider, Farzad Khalvati
2021 Scientific Reports  
We also tested the prognostic performance for overall survival using four feature fusion and reduction methods for combining radiomics and transfer learning features and compared the results with our proposed  ...  AbstractAs an analytic pipeline for quantitative imaging feature extraction and analysis, radiomics has grown rapidly in the past decade.  ...  Acknowledgements This study was conducted with support of the Ontario Institute for Cancer Research (PanCuRx Translational Research Initiative) through funding provided by the Government of Ontario, the  ... 
doi:10.1038/s41598-021-80998-y pmid:33446870 fatcat:iqpkiba725g7vgdw3nc22tmfpu
« Previous Showing results 1 — 15 out of 447 results