TIGERr: Technical Variation Elimination with Ensemble Learning Architecture

The R implementation of TIGER. TIGER integrates random forest algorithm into an innovative ensemble learning architecture. Benefiting from this advanced architecture, TIGER is resilient to outliers, free from model tuning and less likely to be affected by specific hyperparameters. TIGER supports targeted and untargeted metabolomics data and is competent to perform both intra- and inter-batch technical variation removal. TIGER can also be used for cross-kit adjustment to ensure data obtained from different analytical assays can be effectively combined and compared. Reference: Han S. et al. (2022) <doi:10.1093/bib/bbab535>.

Version: 1.0.0
Depends: R (≥ 3.5.0)
Imports: parallel (≥ 2.1.0), pbapply (≥ 1.4-3), ppcor (≥ 1.1), randomForest (≥ 4.6-14), stats (≥ 3.0.0)
Published: 2022-01-06
Author: Siyu Han [aut, cre], Jialing Huang [aut], Francesco Foppiano [aut], Cornelia Prehn [aut], Jerzy Adamski [aut], Karsten Suhre [aut], Ying Li [aut], Giuseppe Matullo [aut], Freimut Schliess [aut], Christian Gieger [aut], Annette Peters [aut], Rui Wang-Sattler [aut]
Maintainer: Siyu Han <siyu.han at helmholtz-muenchen.de>
BugReports: https://github.com/HAN-Siyu/TIGER/issues
License: GPL (≥ 3)
NeedsCompilation: no
Citation: TIGERr citation info
Materials: NEWS
CRAN checks: TIGERr results

Documentation:

Reference manual: TIGERr.pdf

Downloads:

Package source: TIGERr_1.0.0.tar.gz
Windows binaries: r-devel: TIGERr_1.0.0.zip, r-release: TIGERr_1.0.0.zip, r-oldrel: TIGERr_1.0.0.zip
macOS binaries: r-release (arm64): TIGERr_1.0.0.tgz, r-oldrel (arm64): TIGERr_1.0.0.tgz, r-release (x86_64): TIGERr_1.0.0.tgz
Old sources: TIGERr archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=TIGERr to link to this page.