Korbinian Strimmer

The "binda" package implements functions for multi-class discriminant analysis using binary predictors, for corresponding variable selection, and for dichotomizing continuous data.

Implements the regression approach of Zuber and Strimmer (2011) "High-dimensional regression and variable selection using CAR scores" SAGMB 10: 34. CAR scores measure the correlation between the response and the Mahalanobis-decorrelated predictors. The squared CAR score is a natural measure of variable importance and provides a canonical ordering of variables. This package provides functions for estimating CAR scores, for variable selection using CAR scores, and for estimating corresponding regression coefficients. Both shrinkage as well as empirical estimators are available.

Implements a James-Stein-type shrinkage estimator for the covariance matrix, with separate shrinkage for variances and correlations. The details of the method are explained in Schafer and Strimmer (2005) <DOI:10.2202/1544-6115.1175> and Opgen-Rhein and Strimmer (2007) <DOI:10.2202/1544-6115.1252>. The approach is both computationally as well as statistically very efficient, it is applicable to "small n, large p" data, and always returns a positive definite and well-conditioned covariance matrix. In addition to inferring the covariance matrix the package also provides shrinkage estimators for partial correlations and partial variances. The inverse of the covariance and correlation matrix can be efficiently computed, as well as any arbitrary power of the shrinkage correlation matrix. Furthermore, functions are available for fast singular value decomposition, for computing the pseudoinverse, and for checking the rank and positive definiteness of a matrix.

Contains generic functions for performing cross validation and for computing diagnostic errors.

This package implements various estimators of entropy, such as the shrinkage estimator by Hausser and Strimmer, the maximum likelihood and the Millow-Madow estimator, various Bayesian estimators, and the Chao-Shen estimator. It also offers an R interface to the NSB estimator. Furthermore, it provides functions for estimating Kullback-Leibler divergence, chi-squared, mutual information, and chi-squared statistic of independence. In addition there are functions for discretizing continuous random variables.

Estimates both tail area-based false discovery rates (Fdr) as well as local false discovery rates (fdr) for a variety of null models (p-values, z-scores, correlation coefficients, t-scores). The proportion of null values and the parameters of the null distribution are adaptively estimated from the data. In addition, the package contains functions for non-parametric density estimation (Grenander estimator), for monotone regression (isotonic regression and antitonic regression with weights), for computing the greatest convex minorant (GCM) and the least concave majorant (LCM), for the half-normal and correlation distributions, and for computing empirical higher criticism (HC) scores and the corresponding decision threshold.

Analyzes gene expression (time series) data with focus on the inference of gene networks. In particular, GeneNet implements the methods of Schaefer and Strimmer (2005a,b,c) and Opgen-Rhein and Strimmer (2006, 2007) for learning large-scale gene association networks (including assignment of putative directions).

Contains general data structures and functions for longitudinal data with multiple variables, repeated measurements, and irregularly spaced time points. Also implements a shrinkage estimator of dynamical correlation and dynamical covariance.

Provides an efficient framework for high-dimensional linear and diagonal discriminant analysis with variable selection. The classifier is trained using James-Stein-type shrinkage estimators and predictor variables are ranked using correlation-adjusted t-scores (CAT scores). Variable selection error is controlled using false non-discovery rates or higher criticism.

Implements the "shrinkage t" statistic introduced in Opgen-Rhein and Strimmer (2007) and a shrinkage estimate of the "correlation-adjusted t-score" (CAT score) described in Zuber and Strimmer (2009). It also offers a convenient interface to a number of other regularized t-statistics commonly employed in high-dimensional case-control studies.

Implements the whitening methods (ZCA, PCA, Cholesky, ZCA-cor, and PCA-cor) discussed in Kessy, Lewin, and Strimmer (2018) "Optimal whitening and decorrelation", <doi:10.1080/00031305.2016.1277159>, as well as the whitening approach to canonical correlation analysis allowing negative canonical correlations described in Jendoubi and Strimmer (2019) "A whitening approach to probabilistic canonical correlation analysis for omics data integration", <doi:10.1186/s12859-018-2572-9>.

Functions for reading, writing, plotting, and manipulating phylogenetic trees, analyses of comparative data in a phylogenetic framework, ancestral character analyses, analyses of diversification and macroevolution, computing distances from DNA sequences, reading and writing nucleotide sequences as well as importing from BioConductor, and several tools such as Mantel's test, generalized skyline plots, graphical exploration of phylogenetic data (alex, trex, kronoviz), estimation of absolute evolutionary rates and clock-like trees using mean path lengths and penalized likelihood, dating trees with non-contemporaneous sequences, translating DNA into AA sequences, and assessing sequence alignments. Phylogeny estimation can be done with the NJ, BIONJ, ME, MVR, SDM, and triangle methods, and several methods handling incomplete distance matrices (NJ*, BIONJ*, MVR*, and the corresponding triangle method). Some functions call external applications (PhyML, Clustal, T-Coffee, Muscle) whose results are returned into R.

The GeneCycle package implements the approaches of Wichert et al. (2004) <doi.org/10.1093/bioinformatics/btg364>, Ahdesmaki et al. (2005) <DOI:10.1186/1471-2105-6-117> and Ahdesmaki et al. (2007) <DOI:10.1186/1471-2105-8-233> for detecting periodically expressed genes from gene expression time series data.

Computation of local false discovery rates.

A complete analysis pipeline for matrix-assisted laser desorption/ionization-time-of-flight (MALDI-TOF) and other two-dimensional mass spectrometry data. In addition to commonly used plotting and processing methods it includes distinctive features, namely baseline subtraction methods such as morphological filters (TopHat) or the statistics-sensitive non-linear iterative peak-clipping algorithm (SNIP), peak alignment using warping functions, handling of replicated measurements as well as allowing spectra with different resolutions.

Routines for PLS-based genomic analyses, implementing PLS methods for classification with microarray data and prediction of transcription factor activities from combined ChIP-chip analysis. The >=1.2-1 versions include two new classification methods for microarray data: GSIM and Ridge PLS. The >=1.3 versions includes a new classification method combining variable selection and compression in logistic regression context: logit-SPLS; and an adaptive version of the sparse PLS.