OPLS discriminant analysis: combining the strengths of PLS‐DA and SIMCA classification

Abstract

The characteristics of the OPLS method have been investigated for the purpose of discriminant analysis (OPLS-DA). We demonstrate how class-orthogonal variation can be exploited to augment classification performance in cases where the individual classes exhibit divergence in within-class variation, in analogy with soft independent modelling of class analogy (SIMCA) classification. The prediction results will be largely equivalent to traditional supervised classification using PLS-DA if no such variation is present in the classes. A discriminatory strategy is thus outlined, combining the strengths of PLS-DA and SIMCA classification within the framework of the OPLS-DA method. Furthermore, resampling methods have been employed to generate distributions of predicted classification results and subsequently assess classification belief. This enables utilisation of the class-orthogonal variation in a proper statistical context. The proposed decision rule is compared to common decision rules and is shown to produce comparable or less class-biased classification results.

Chemometrics in Metabonomics

“…), it is recommended to use other supervised methods such as PLS-DA in the two-class cases (or multiple class cases) and/or preferably OPLS-DA 19 to facilitate the interpretation ( Figure 5). …”

UPLC-based metabonomic applications for discovering biomarkers of diseases in clinical chemistry

“…The OPLS-DA score plot revealed good fitness and high predictability of the OPLS-DA model with high statistical values of R 2 and Q 2 [23]. OPLS-DA method was used as a complement to the PLS-DA to discriminate two or more groups using multivariate data [24,25]. OPLS-DA model is calculated between the multivariate data and a response variable that only contains class information.…”

Metabolic profiling reveals disorder of amino acid metabolism in four brain regions from a rat model of chronic unpredictable mild stress

“…Multivariate statistics, including unsupervised PCA and supervised orthogonal partial least squares project to latent structures-discriminant analysis (OPLS-DA), was performed by SIMCA-P 11.0 software (Umetrics, Umeå, Sweden) [34,35]. The data set was mean-centered and pareto-scaled in a columnwise manner for all the multivariate modeling [36].…”

“…The resulting PCA scores map was used for investigating natural interrelation including possible groupings, clustering, and outliers among observations. Furthermore, a more sophisticated OPLS-DA model was achieved through removing the variation in X matrix unrelated to Y matrix so that the specific discriminant information between classes can be interpreted using one predictive component alone [34,37]. Great efforts have been made to test the reliability of multivariate models [38,39], hence the 6-round cross-validation in SMICA-P software was herein applied to validate the OPLS-DA model against over-fitting by precluding 1/6th of all the samples in each round.…”