ropls: PCA, PLS(-DA) and OPLS(-DA) for multivariate analysis and feature selection of omics data
2 May 2019
Package
ropls 1.16.0
Contents
- 1 The ropls package
- 2 Context
- 3 The sacurine metabolomics dataset
- 4 Hands-on
- 5 Other datasets
- 6 Session info
- References
1 The ropls package
The ropls R package implements the PCA, PLS(-DA) and OPLS(-DA) approaches with the original, NIPALS-based, versions of the algorithms (Wold, Sjostrom, and Eriksson 2001, Trygg and Wold (2002)). It includes the R2 and Q2 quality metrics (Eriksson et al. 2001, Tenenhaus (1998)), the permutation diagnostics (Szymanska et al. 2012), the computation of the VIP values (Wold, Sjostrom, and Eriksson 2001), the score and orthogonal distances to detect outliers (Hubert, Rousseeuw, and Vanden Branden 2005), as well as many graphics (scores, loadings, predictions, diagnostics, outliers, etc).
The functionalities from ropls can also be accessed via a graphical user interface in the Multivariate module from the Workflow4Metabolomics.org online resource for computational metabolomics, which provides a user-friendly, Galaxy-based environment for data pre-processing, statistical analysis, and annotation (Giacomoni et al. 2015).
2 Context
2.1 Orthogonal Partial Least-Squares
Partial Least-Squares (PLS), which is a latent variable regression method based on covariance between the predictors and the response, has been shown to efficiently handle datasets with multi-collinear predictors, as in the case of spectrometry measurements (Wold, Sjostrom, and Eriksson 2001). More recently, Trygg and Wold (2002) introduced the Orthogonal Partial Least-Squares (OPLS) algorithm to model separately the variations of the predictors correlated and orthogonal to the response.
OPLS has a similar predictive capacity compared to PLS and improves the interpretation of the predictive components and of the systematic variation (Pinto, Trygg, and Gottfries 2012). In particular, OPLS modeling of single responses only requires one predictive component.
Diagnostics such as the Q2Y metrics and permutation testing are of high importance to avoid overfitting and assess the statistical significance of the model. The Variable Importance in Projection (VIP), which reflects both the loading weights for each component and the variability of the response explained by this component (Pinto, Trygg, and Gottfries 2012; Mehmood et al. 2012), can be used for feature selection (Trygg and Wold 2002; Pinto, Trygg, and Gottfries 2012).
2.2 OPLS software
OPLS is available in the SIMCA-P commercial software (Umetrics, Umea, Sweden; Eriksson et al. (2001)). In addition, the kernel-based version of OPLS (Bylesjo et al. 2008) is available in the open-source R statistical environment (R Development Core Team 2008), and a single implementation of the linear algorithm in R has been described recently (Gaude et al. 2013).
3 The sacurine metabolomics dataset
3.1 Study objective
The objective was to study the influence of age, body mass index (bmi), and gender on metabolite concentrations in urine, by analysing 183 samples from a cohort of adults with liquid chromatography coupled to high-resolution mass spectrometry (LC-HRMS; Thevenot et al. (2015)).
3.2 Pre-processing and annotation
Urine samples were analyzed by using an LTQ Orbitrap in the negative ionization mode. A total of 109 metabolites were identified or annotated at the MSI level 1 or 2. After retention time alignment with XCMS, peaks were integrated with Quan Browser. Signal drift and batch effect were corrected, and each urine profile was normalized to the osmolality of the sample. Finally, the data were log10 transformed (Thevenot et al. 2015).
3.3 Covariates
The volunteers’ age, body mass index (bmi), and gender were recorded.
4 Hands-on
4.1 Loading
We first load the ropls package:
library(ropls)We then load the sacurine dataset which contains:
The dataMatrix matrix of numeric type containing the intensity profiles (log10 transformed),
The sampleMetadata data frame containg sample metadata,
The variableMetadata data frame containg variable metadata
data(sacurine)
names(sacurine)## [1] "dataMatrix" "sampleMetadata" "variableMetadata"We attach sacurine to the search path and display a summary of the content
of the dataMatrix, sampleMetadata and variableMetadata with the
strF Function of the
ropls package
(see also str):
attach(sacurine)strF(dataMatrix)## dim class mode typeof size NAs min mean median max
## 183 x 109 matrix numeric double 0.2 Mb 0 -0.3 4.2 4.3 6
## (2-methoxyethoxy)propanoic acid isomer (gamma)Glu-Leu/Ile ...
## HU_011 3.019766011 3.888479324 ...
## HU_014 3.81433889 4.277148905 ...
## ... ... ... ...
## HU_208 3.748127215 4.523763202 ...
## HU_209 4.208859398 4.675880567 ...
## Valerylglycine isomer 2 Xanthosine
## HU_011 3.889078716 4.075879575
## HU_014 4.181765852 4.195761901
## ... ... ...
## HU_208 4.634338821 4.487781609
## HU_209 4.47194762 4.222953354strF(sampleMetadata)## age bmi gender
## numeric numeric factor
## nRow nCol size NAs
## 183 3 0 Mb 0
## age bmi gender
## HU_011 29 19.75 M
## HU_014 59 22.64 F
## ... ... ... ...
## HU_208 27 18.61 F
## HU_209 17.5 21.48 FstrF(variableMetadata)## msiLevel hmdb chemicalClass
## numeric character character
## nRow nCol size NAs
## 109 3 0 Mb 0
## msiLevel hmdb chemicalClass
## (2-methoxyethoxy)propanoic acid isomer 2 Organi
## (gamma)Glu-Leu/Ile 2 AA-pep
## ... ... ... ...
## Valerylglycine isomer 2 2 AA-pep:AcyGly
## Xanthosine 1 HMDB00299 NucleoWe may also have a look at the data matrix with the imageF function:
imageF(dataMatrix)
4.2 Principal Component Analysis (PCA)
We perform a PCA on the dataMatrix matrix (samples as rows, variables as
columns), with the opls method:
sacurine.pca <- opls(dataMatrix)A summary of the model (8 components were selected) is printed:
## PCA
## 183 samples x 109 variables
## standard scaling of predictors
## R2X(cum) pre ort
## Total 0.501 8 0In addition the default summary figure is displayed:
Figure 1: PCA summary plot. Top left overview: the scree plot (i.e.,
inertia barplot) suggests that 3 components may be sufficient to capture most of
the inertia; Top right outlier: this graphics shows the distances within
and orthogonal to the projection plane (Hubert, Rousseeuw, and Vanden Branden 2005): the name of the samples
with a high value for at least one of the distances are indicated; Bottom
left x-score: the variance along each axis equals the variance captured by
each component: it therefore depends on the total variance of the dataMatrix X
and of the percentage of this variance captured by the component (indicated in
the labels); it decreases when going from one component to a component with
higher indice; Bottom right x-loading: the 3 variables with most extreme
values (positive and negative) for each loading are black colored and labeled.
Note:
Since dataMatrix does not contain missing value, the singular value decomposition was used by default; NIPALS can be selected with the
algoCargument specifying the algorithm (Character),The
predI = NAdefault number of predictive components (Integer) for PCA means that components (up to 10) will be computed until the cumulative variance exceeds 50%. If the 50% have not been reached at the 10th component, a warning message will be issued (you can still compute the following components by specifying thepredIvalue).
Let us see if we notice any partition according to gender, by labeling/coloring
the samples according to gender (parAsColFcVn) and drawing the Mahalanobis
ellipses for the male and female subgroups (parEllipseL).
genderFc <- sampleMetadata[, "gender"]
plot(sacurine.pca,
typeVc = "x-score",
parAsColFcVn = genderFc)
Figure 2: PCA score plot colored according to gender.
Note:
The plotting parameter to be used As Colors (Factor of character type or Vector of numeric type) has a length equal to the number of rows of the dataMatrix (ie of samples) and that this qualitative or quantitative variable is converted into colors (by using an internal palette or color scale, respectively). We could have visualized the age of the individuals by specifying
parAsColFcVn = sampleMetadata[, "age"].The displayed components can be specified with
parCompVi(plotting parameter specifying the Components: Vector of 2 integers)The labels and the color palette can be personalized with the
parLabVcandparPaletteVcparameters, respectively:
plot(sacurine.pca,
typeVc = "x-score",
parAsColFcVn = genderFc,
parLabVc = as.character(sampleMetadata[, "age"]),
parPaletteVc = c("green4", "magenta"))
4.3 Partial least-squares: PLS and PLS-DA
For PLS (and OPLS), the Y response(s) must be provided to the opls
method. Y can be either a numeric vector (respectively matrix) for single
(respectively multiple) (O)PLS regression, or a character factor for
(O)PLS-DA classification as in the following example with the gender
qualitative response:
sacurine.plsda <- opls(dataMatrix, genderFc)## PLS-DA
## 183 samples x 109 variables and 1 response
## standard scaling of predictors and response(s)
## R2X(cum) R2Y(cum) Q2(cum) RMSEE pre ort pR2Y pQ2
## Total 0.275 0.73 0.584 0.262 3 0 0.05 0.05
Figure 3: PLS-DA model of the gender response. Top left:
significance diagnostic: the R2Y and Q2Y of the model are compared
with the corresponding values obtained after random permutation of the y
response; Top right: inertia barplot: the graphic here suggests that 3
orthogonal components may be sufficient to capture most of the inertia; Bottom
left: outlier diagnostics; Bottom right: x-score plot: the number of
components and the cumulative R2X, R2Y and Q2Y are indicated below the
plot.
Note:
- When set to NA (as in the default), the number of components
predIis determined automatically as follows (Eriksson et al. 2001): A new component h is added to the model if:
\(R2Y_h \geq 0.01\), i.e., the percentage of Y dispersion (i.e., sum of squares) explained by component h is more than 1 percent, and
\(Q2Y_h=1-PRESS_h/RSS_{h-1} \geq 0\) for PLS (or 5% when the number of samples is less than 100) or 1% for OPLS: \(Q2Y_h \geq 0\) means that the predicted residual sum of squares (\(PRESS_h\)) of the model including the new component h estimated by 7-fold cross-validation is less than the residual sum of squares (\(RSS_{h-1}\)) of the model with the previous components only (with \(RSS_0\) being the sum of squared Y values).
The predictive performance of the full model is assessed by the cumulative Q2Y metric: \(Q2Y=1-\prod\limits_{h=1}^r (1-Q2Y_h)\). We have \(Q2Y \in [0,1]\), and the higher the Q2Y, the better the performance. Models trained on datasets with a larger number of features compared with the number of samples can be prone to overfitting: in that case, high Q2Y values are obtained by chance only. To estimate the significance of Q2Y (and R2Y) for single response models, permutation testing (Szymanska et al. 2012) can be used: models are built after random permutation of the Y values, and \(Q2Y_{perm}\) are computed. The p-value is equal to the proportion of \(Q2Y_{perm}\) above \(Q2Y\) (the default number of permutations is 20 as a compromise between quality control and computation speed; it can be increased with the
permIparameter, e.g. to 1,000, to assess if the model is significant at the 0.05 level),The NIPALS algorithm is used for PLS (and OPLS); dataMatrix matrices with (a moderate amount of) missing values can thus be analysed.
We see that our model with 3 predictive (pre) components has significant and quite high R2Y and Q2Y values.
4.4 Orthogonal partial least squares: OPLS and OPLS-DA
To perform OPLS(-DA), we set orthoI (number of components which are
orthogonal; Integer) to either a specific number of orthogonal components,
or to NA. Let us build an OPLS-DA model of the gender response.
sacurine.oplsda <- opls(dataMatrix, genderFc,
predI = 1, orthoI = NA)## OPLS-DA
## 183 samples x 109 variables and 1 response
## standard scaling of predictors and response(s)
## R2X(cum) R2Y(cum) Q2(cum) RMSEE pre ort pR2Y pQ2
## Total 0.275 0.73 0.602 0.262 1 2 0.05 0.05
Figure 4: OPLS-DA model of the gender response.
Note:
For OPLS modeling of a single response, the number of predictive component is 1,
In the (
x-scoreplot), the predictive component is displayed as abscissa and the (selected; default = 1) orthogonal component as ordinate.
Let us assess the predictive performance of our model. We first train the
model on a subset of the samples (here we use the odd subset value which
splits the data set into two halves with similar proportions of samples for each
class; alternatively, we could have used a specific subset of indices for
training):
sacurine.oplsda <- opls(dataMatrix, genderFc,
predI = 1, orthoI = NA,
subset = "odd")## OPLS-DA
## 92 samples x 109 variables and 1 response
## standard scaling of predictors and response(s)
## R2X(cum) R2Y(cum) Q2(cum) RMSEE RMSEP pre ort
## Total 0.26 0.825 0.608 0.213 0.341 1 2
We first check the predictions on the training subset:
trainVi <- getSubsetVi(sacurine.oplsda)
table(genderFc[trainVi], fitted(sacurine.oplsda))##
## M F
## M 50 0
## F 0 42We then compute the performances on the test subset:
table(genderFc[-trainVi],
predict(sacurine.oplsda, dataMatrix[-trainVi, ]))##
## M F
## M 43 7
## F 7 34As expected, the predictions on the test subset are (slightly) lower. The classifier however still achieves 91% of correct predictions.
4.6 Working on ExpressionSet omics objects from bioconductor
The ExpressionSet class from the
Biobase
bioconductor package has been developed to conveniently handle preprocessed
omics objects, including the variables x samples matrix of intensities, and data
frames containing the sample and variable metadata (Huber et al. 2015). The matrix and
the two data frames can be accessed by the exprs, pData and fData
respectively (note that the data matrix is stored in the object with samples in
columns).
The opls method can be applied to an ExpressionSet object, by using the
object as the x argument, and, for (O)PLS(-DA), by indicating as the y
argument the name of the sampleMetadata to be used as the response.
In the example below, we will first build a minimal ExpressionSet object from the sacurine data set, and we subsequently perform an OPLS-DA.
library(Biobase)
sacSet <- ExpressionSet(assayData = t(dataMatrix),
phenoData = new("AnnotatedDataFrame",
data = sampleMetadata))
opls(sacSet, "gender", orthoI = NA)## OPLS-DA
## 183 samples x 109 variables and 1 response
## standard scaling of predictors and response(s)
## R2X(cum) R2Y(cum) Q2(cum) RMSEE pre ort pR2Y pQ2
## Total 0.275 0.73 0.602 0.262 1 2 0.05 0.05
4.7 Importing/exporting data from/to the Workflow4metabolomics infrastructure
Galaxy is a web-based environment providing powerful graphical user interface and workflow management functionalities for omics data analysis (Goecks et al. (2010); Boekel et al. (2015)). Wrapping an R code into a Galaxy module is quite straight-forward: examples can be found on the toolshed central repository and in the RGalaxy bioconductor package.
Workflow4metabolomics (W4M) is the online
infrastructure for computational metabolomics based on the Galaxy environment
(Giacomoni et al. 2015). W4M enables to build, run, save and share workflows
efficiently. In addition, workflows and input/output data (called histories)
can be
referenced,
thus enabling fully reproducible research. More than 30 modules are currently
available for LC-MS, GC-MS and NMR data preprocessing, statistical analysis, and
annotation, including wrappers of xcms, CAMERA, metaMS, ropls, and
biosigner, and is open to
new contributions.
In order to facilitate data import from/to W4M, the fromW4M function
(respectively the toW4M method) enables import from (respectively export to)
the W4M 3 tabular file format (dataMatrix.tsv, sampleMetadata.tsv,
variableMetadata.tsv) into (respectively from) an ExpressionSet object, as
shown in the following example which uses the 3 .tsv files stored in the
extdata repository of the package to create a sacSet ExpressionSet object:
sacSet <- fromW4M(file.path(path.package("ropls"), "extdata"))
sacSet## ExpressionSet (storageMode: lockedEnvironment)
## assayData: 109 features, 183 samples
## element names: exprs
## protocolData: none
## phenoData
## sampleNames: HU_011 HU_014 ... HU_209 (183 total)
## varLabels: age bmi gender
## varMetadata: labelDescription
## featureData
## featureNames: X.2.methoxyethoxy.propanoic.acid.isomer
## X.gamma.Glu.Leu.Ile ... Xanthosine (109 total)
## fvarLabels: msiLevel hmdb chemicalClass
## fvarMetadata: labelDescription
## experimentData: use 'experimentData(object)'
## Annotation:The generated sacSet ExpressionSet object can be used with the opls method
as described in the previous section.
Conversely, an ExpressionSet (with filled exprs, phenoData and
featureData slots) can be exported to the 3 table W4M format:
toW4M(sacSet, paste0(getwd(), "/out_"))Before moving to the next session whith another example dataset, we detach sacurine from the search path:
detach(sacurine)5 Other datasets
In addition to the sacurine dataset presented above, the package contains the following datasets to illustrate the functionalities of PCA, PLS and OPLS (see the examples in the documentation of the opls function):
aminoacids Amino-Acids Dataset. Quantitative structure property relationship (QSPR) (Wold, Sjostrom, and Eriksson 2001).
cellulose NIR-Viscosity example data set to illustrate multivariate calibration using PLS, spectral filtering and OPLS (Multivariate calibration using spectral data. Simca tutorial. Umetrics, Sweden).
cornell Octane of various blends of gasoline: Twelve mixture component proportions of the blend are analysed (Tenenhaus 1998).
foods Food consumption patterns accross European countries (FOODS). The relative consumption of 20 food items was compiled for 16 countries. The values range between 0 and 100 percent and a high value corresponds to a high consumption. The dataset contains 3 missing data (Eriksson et al. 2001).
linnerud Three physiological and three exercise variables are measured on twenty middle-aged men in a fitness club (Tenenhaus 1998).
lowarp A multi response optimization data set (LOWARP) (Eriksson et al. 2001).
mark Marks obtained by french students in mathematics, physics, french and english. Toy example to illustrate the potentialities of PCA (Baccini 2010).
6 Session info
Here is the output of sessionInfo() on the system on which this document was
compiled:
## R version 3.6.0 (2019-04-26)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 18.04.2 LTS
##
## Matrix products: default
## BLAS: /home/biocbuild/bbs-3.9-bioc/R/lib/libRblas.so
## LAPACK: /home/biocbuild/bbs-3.9-bioc/R/lib/libRlapack.so
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] parallel stats graphics grDevices utils datasets methods
## [8] base
##
## other attached packages:
## [1] ropls_1.16.0 Biobase_2.44.0 BiocGenerics_0.30.0
## [4] BiocStyle_2.12.0
##
## loaded via a namespace (and not attached):
## [1] Rcpp_1.0.1 bookdown_0.9 digest_0.6.18
## [4] magrittr_1.5 evaluate_0.13 stringi_1.4.3
## [7] rmarkdown_1.12 tools_3.6.0 stringr_1.4.0
## [10] xfun_0.6 yaml_2.2.0 compiler_3.6.0
## [13] BiocManager_1.30.4 htmltools_0.3.6 knitr_1.22References
Baccini, A. 2010. “Statistique Descriptive Multidimensionnelle (Pour Les Nuls).”
Boekel, J., JM. Chilton, IR. Cooke, PL. Horvatovich, PD. Jagtap, L. Kall, J. Lehtio, P. Lukasse, PD. Moerland, and TJ. Griffin. 2015. “Multi-Omic Data Analysis Using Galaxy.” Nature Biotechnology 33 (2):137–39. https://doi.org/10.1038/nbt.3134.
Brereton, Richard G., and Gavin R. Lloyd. 2014. “Partial Least Squares Discriminant Analysis: Taking the Magic Away.” Journal of Chemometrics 28 (4):213–25. http://dx.doi.org/10.1002/cem.2609.
Bylesjo, M., M. Rantalainen, J. Nicholson, E. Holmes, and J. Trygg. 2008. “K-OPLS Package: Kernel-Based Orthogonal Projections to Latent Structures for Prediction and Interpretation in Feature Space.” BMC Bioinformatics 9 (1):106. http://dx.doi.org/10.1186/1471-2105-9-106.
Bylesjo, M, M Rantalainen, O Cloarec, J Nicholson, E Holmes, and J Trygg. 2006. “OPLS Discriminant Analysis: Combining the Strengths of PLS-DA and SIMCA Classification.” Journal of Chemometrics 20:341–51. http://dx.doi.org/10.1002/cem.1006.
Eriksson, L., E. Johansson, N. Kettaneh-Wold, and S. Wold. 2001. Multi- and Megavariate Data Analysis. Principles and Applications. Umetrics Academy.
Galindo-Prieto, B., L. Eriksson, and J. Trygg. 2014. “Variable Influence on Projection (VIP) for Orthogonal Projections to Latent Structures (OPLS).” Journal of Chemometrics 28 (8):623–32. http://dx.doi.org/10.1002/cem.2627.
Gaude, R., F. Chignola, D. Spiliotopoulos, A. Spitaleri, M. Ghitti, JM. Garcia-Manteiga, S. Mari, and G. Musco. 2013. “Muma, an R Package for Metabolomics Univariate and Multivariate Statistical Analysis.” Current Metabolomics 1:180–89. http://dx.doi.org/10.2174/2213235X11301020005.
Giacomoni, F., G. Le Corguille, M. Monsoor, M. Landi, P. Pericard, M. Petera, C. Duperier, et al. 2015. “Workflow4Metabolomics: A Collaborative Research Infrastructure for Computational Metabolomics.” Bioinformatics 31 (9):1493–5. http://dx.doi.org/10.1093/bioinformatics/btu813.
Goecks, J., A. Nekrutenko, J. Taylor, and The Galaxy Team. 2010. “Galaxy: A Comprehensive Approach for Supporting Accessible, Reproducible, and Transparent Computational Research in the Life Sciences.” Genome Biology 11 (8):R86. https://doi.org/10.1186/gb-2010-11-8-r86.
Huber, W., VJ. Carey, R. Gentleman, S. Anders, M. Carlson, BS. Carvalho, HC. Bravo, et al. 2015. “Orchestrating High-Throughput Genomic Analysis with Bioconductor.” Nature Methods 12 (2):115–21. https://doi.org/10.1038/nmeth.3252.
Hubert, M., PJ. Rousseeuw, and K. Vanden Branden. 2005. “ROBPCA: A New Approach to Robust Principal Component Analysis.” Technometrics 47:64–79. http://dx.doi.org/10.1198/004017004000000563.
Mehmood, T., KH. Liland, L. Snipen, and S. Saebo. 2012. “A Review of Variable Selection Methods in Partial Least Squares Regression.” Chemometrics and Intelligent Laboratory Systems 118 (0):62–69. http://dx.doi.org/10.1016/j.chemolab.2012.07.010.
Pinto, RC., J. Trygg, and J. Gottfries. 2012. “Advantages of Orthogonal Inspection in Chemometrics.” Journal of Chemometrics 26 (6):231–35. http://dx.doi.org/10.1002/cem.2441.
R Development Core Team. 2008. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org.
Szymanska, E., E. Saccenti, AK. Smilde, and JA. Westerhuis. 2012. “Double-Check: Validation of Diagnostic Statistics for PLS-DA Models in Metabolomics Studies.” Metabolomics 8 (1, 1):3–16. http://dx.doi.org/10.1007/s11306-011-0330-3.
Tenenhaus, M. 1998. La Regression PLS : Theorie et Pratique. Editions Technip.
Thevenot, EA., A. Roux, X. Ying, E. Ezan, and C. Junot. 2015. “Analysis of the Human Adult Urinary Metabolome Variations with Age, Body Mass Index and Gender by Implementing a Comprehensive Workflow for Univariate and OPLS Statistical Analyses.” Journal of Proteome Research 14 (8):3322–35. http://dx.doi.org/10.1021/acs.jproteome.5b00354.
Trygg, J., and S. Wold. 2002. “Orthogonal Projection to Latent Structures (O-PLS).” Journal of Chemometrics 16:119–28. http://dx.doi.org/10.1002/cem.695.
Wehrens, R. 2011. Chemometrics with R: Multivariate Data Analysis in the Natural Sciences and Life Sciences. Springer.
Wold, S., M. Sjostrom, and L. Eriksson. 2001. “PLS-Regression: A Basic Tool of Chemometrics.” Chemometrics and Intelligent Laboratory Systems 58:109–30. http://dx.doi.org/10.1016/S0169-7439(01)00155-1.
4.5 Comments
4.5.1 Overfitting
Overfitting (i.e., building a model with good performances on the training set but poor performances on a new test set) is a major caveat of machine learning techniques applied to data sets with more variables than samples. A simple simulation of a random X data set and a y response shows that perfect PLS-DA classification can be achieved as soon as the number of variables exceeds the number of samples, as detailed in the example below, adapted from Wehrens (2011):
This simple simulation illustrates that PLS overfit can occur, in particular when the number of features exceeds the number of observations. It is therefore essential to check that the \(Q2Y\) value of the model is significant by random permutation of the labels.
4.5.2 VIP from OPLS models
The classical VIP metric is not useful for OPLS modeling of a single response since (Galindo-Prieto, Eriksson, and Trygg 2014, Thevenot et al. (2015)): 1. VIP values remain identical whatever the number of orthogonal components selected, 2. VIP values are univariate (i.e., they do not provide information about interactions between variables). In fact, when features are standardized, we can demonstrate a mathematical relationship between VIP and p-values from a Pearson correlation test (Thevenot et al. 2015), as illustrated by the figure below:
Figure 6: Relationship between VIP from one-predictive PLS or OPLS models with standardized variables, and p-values from Pearson correlation test. The \((p_j, VIP_j)\) pairs corresponding respectively to the VIP values from OPLS modelling of the age response with the sacurine dataset, and the p-values from the Pearson correlation test are shown as red dots. The \(y = \Phi^{-1}(1 - x/2) / z_{rms}\) curve is shown in red (where \(\Phi^{-1}\) is the inverse of the probability density function of the standard normal distribution, and \(z_{rms}\) is the quadratic mean of the \(z_j\) quantiles from the standard normal distribution; \(z_{rms} = 2.6\) for the sacurine dataset and the age response). The vertical (resp. horizontal) blue line corresponds to univariate (resp. multivariate) thresholds of \(p=0.05\) and \(VIP=1\), respectively (Thevenot et al. 2015).
The VIP properties above result from:
OPLS models of a single response have a single predictive component,
in the case of one-predictive component (O)PLS models, the general formula for VIPs can be simplified to \(VIP_j = \sqrt{m} \times |w_j|\) for each feature \(j\), were \(m\) is the total number of features and w is the vector of loading weights,
in OPLS, w remains identical whatever the number of extracted orthogonal components,
for a single-response model, w is proportional to X’y (where ’ denotes the matrix transposition),
if X and y are standardized, X’y is the vector of the correlations between the features and the response.
Galindo-Prieto, Eriksson, and Trygg (2014) have recently suggested new VIP metrics for OPLS, VIP_pred and VIP_ortho, to separately measure the influence of the features in the modeling of the dispersion correlated to, and orthogonal to the response, respectively (Galindo-Prieto, Eriksson, and Trygg 2014).
For OPLS(-DA) models, you can therefore get from the model generated with
opls:the predictive VIP vector (which corresponds to the \(VIP_{4,pred}\) metric measuring the variable importance in prediction) with
getVipVn(model),the orthogonal VIP vector which is the \(VIP_{4,ortho}\) metric measuring the variable importance in orthogonal modeling with
getVipVn(model, orthoL = TRUE). As for the classical VIP, we still have the mean of \(VIP_{pred}^2\) (and of \(VIP_{ortho}^2\)) which, each, equals 1.4.5.3 (Orthogonal) Partial Least Squares Discriminant Analysis: (O)PLS-DA
4.5.3.1 Two classes
When the y response is a factor of 2 levels (character vectors are also allowed), it is internally transformed into a vector of values \(\in \{0,1\}\) encoding the classes. The vector is centered and unit-variance scaled, and the (O)PLS analysis is performed.
Brereton and Lloyd (2014) have demonstrated that when the sizes of the 2 classes are unbalanced, a bias is introduced in the computation of the decision rule, which penalizes the class with the highest size (Brereton and Lloyd 2014). In this case, an external procedure using resampling (to balance the classes) and taking into account the class sizes should be used for optimal results.
4.5.3.2 Multiclass
In the case of more than 2 levels, the y response is internally transformed into a matrix (each class is encoded by one column of values \(\in \{0,1\}\)). The matrix is centered and unit-variance scaled, and the PLS analysis is performed.
In this so-called PLS2 implementation, the proportions of 0 and 1 in the columns is usually unbalanced (even in the case of balanced size of the classes) and the bias described previously occurs (Brereton and Lloyd 2014). The multiclass PLS-DA results from ropls are therefore indicative only, and we recommend to set an external procedure where each column of the matrix is modeled separately (as described above) and the resulting probabilities are aggregated (see for instance Bylesjo et al. (2006)).