The pls package implements principal component regression (PCR) and partial least squares regression (PLSR) in R (R Development Core Team 2006b), and is freely available from the Comprehensive R Archive Network (CRAN), licensed under the GNU General Public License (GPL).The user interface is modelled after the traditional formula interface, as exemplified by lm. This was done so that people used to R would not have to learn yet another interface, and also because we believe the formula interface is a good way of working interactively with models. It thus has methods for generic functions like predict, update and coef. It also has more specialised functions like scores, loadings and RMSEP, and a flexible crossvalidation system. Visual inspection and assessment is important in chemometrics, and the pls package has a number of plot functions for plotting scores, loadings, predictions, coefficients and RMSEP estimates.The package implements PCR and several algorithms for PLSR. The design is modular, so that it should be easy to use the underlying algorithms in other functions. It is our hope that the package will serve well both for interactive data analysis and as a building block for other functions or packages using PLSR or PCR.We will here describe the package and how it is used for data analysis, as well as how it can be used as a part of other packages. Also included is a section about formulas and data frames, for people not used to the R modelling idioms.

In this age of ever-increasing data set sizes, especially in the natural sciences, visualisation becomes more and more important. Self-organizing maps have many features that make them attractive in this respect: they do not rely on distributional assumptions, can handle huge data sets with ease, and have shown their worth in a large number of applications. In this paper, we highlight the kohonen package for R, which implements self-organizing maps as well as some extensions for supervised pattern recognition and data fusion.

No abstract

IntroductionBatch effects in large untargeted metabolomics experiments are almost unavoidable, especially when sensitive detection techniques like mass spectrometry (MS) are employed. In order to obtain peak intensities that are comparable across all batches, corrections need to be performed. Since non-detects, i.e., signals with an intensity too low to be detected with certainty, are common in metabolomics studies, the batch correction methods need to take these into account. ObjectivesThis paper aims to compare several batch correction methods, and investigates the effect of different strategies for handling non-detects.MethodsBatch correction methods usually consist of regression models, possibly also accounting for trends within batches. To fit these models quality control samples (QCs), injected at regular intervals, can be used. Also study samples can be used, provided that the injection order is properly randomized. Normalization methods, not using information on batch labels or injection order, can correct for batch effects as well. Introducing two easy-to-use quality criteria, we assess the merits of these batch correction strategies using three large LC–MS and GC–MS data sets of samples from Arabidopsis thaliana.ResultsThe three data sets have very different characteristics, leading to clearly distinct behaviour of the batch correction strategies studied. Explicit inclusion of information on batch and injection order in general leads to very good corrections; when enough QCs are available, also general normalization approaches perform well. Several approaches are shown to be able to handle non-detects—replacing them with very small numbers such as zero seems the worst of the approaches considered.ConclusionThe use of quality control samples for batch correction leads to good results when enough QCs are available. If an experiment is properly set up, batch correction using the study samples usually leads to a similar high-quality correction, but has the advantage that more metabolites are corrected. The strategy for handling non-detects is important: choosing small values like zero can lead to suboptimal batch corrections.