Compared application of the new OPLS-DA statistical model versus partial least squares regression to manage large numbers of variables in an injury case-control study

Abstract

The use of modern statistical methodology to overcome the known pitfalls of classical regression models in the analysis of large numbers of highly correlated data, has increased considerably in recent years. Statisticians in the field of chemometrics and OMICS research have developed a new method called Orthogonal projections to latent structures (OPLS). In comparison with the regular partial least squares (PLS) regression, OPLS provides a simpler method with the additional advantage that the orthogonal variation can be analyzed separately. Use of the OPLS model has spread to fields other than its origin but it is not yet applied to the field of epidemiology, which is a wide field of research. In public health and clinical research, there are situations in which large numbers of correlated variables need to be modeled. The authors successfully applied OPLS-DA to model large numbers of variables in a casecontrol study and compared it with discriminant analysis done by partial least squares regression. Prior to fitting the models, the dataset was split into two parts: a training set and a prediction set. Models fitted on the training dataset were later tested for validity in the prediction dataset. The OPLS-DA was compared with PLS-DA for model fitness, diagnostics and model interpretability. Both models suited the data but OPLS-DA was preferable. The authors encourage the use of these methods to increase study power and statistical validity in epidemiology and similar settings in which large numbers of correlated variables need to be modeled.

Differences in the metabolomic signatures of porcine follicular fluid collected from environments associated with good and poor oocyte quality

“…The overall quality of the models was judged by cumulative R 2 and the predictive ability by cumulative Q 2 extracted according to the internal cross-validation default method of SIMCA-P C Software. An orthogonal PLS-DA (OPLS-DA) as a supervised model of classes, was also performed using SIMCA-P C Software (Cloarec et al 2005, Eriksson et al 2008, Sadeghi-Bazargani et al 2011. Data were scaled using unit Par scaling prior to OPLS-DA.…”

Facing obesity in pain rehabilitation clinics: Profiles of physical activity in patients with chronic pain and obesity—A study from the Swedish Quality Registry for Pain Rehabilitation (SQRP)

“…However, it is plausible to consider confirmatory analyses because the risks for intercorrelations among the covariates still could exist. OPLS regressions have higher sensitivity, properly deal with multicollinearity, compensate for missing data, and provide higher statistical power [45,46,70]. However, OPLS cannot determine the independent effect of a certain variable when controlling for other variables.…”

“…The logistic regressions suffer some disadvantages, such as assumptions of variable independence, meeting power, and missing data [44][45][46]. Therefore, we also performed confirmatory analyses using advanced Principal Component Analysis (PCA) for the multivariate correlation analyses to detect outliers and Orthogonal Partial Least Square Regressions (OPLS) for the multivariate regressions.…”

Classification of patients with lithium-treated bipolar disorder based on gene expression: Dirichlet Bayesian network model

“…This method is a supervised method that was developed to solve the prediction problem in multivariate problems. This study used the PLS regression and VIP index to select essential variables (19). After choosing the necessary variables to determine the homogeneous and identical patterns in the selected genes, we used the biclustering method, and the representative of each cluster was discovered based on the principal component analysis method.…”