1 Introduction

High-throughput, non-targeted, technologies such as transcriptomics, proteomics and metabolomics, are widely used to discover molecules which allow to efficiently discriminate between biological or clinical conditions of interest (e.g., disease vs control states). Powerful machine learning approaches such as Partial Least Square Discriminant Analysis (PLS-DA), Random Forest (RF) and Support Vector Machines (SVM) have been shown to achieve high levels of prediction accuracy. Feature selection, i.e., the selection of the few features (i.e., the molecular signature) which are of highest discriminating value, is a critical step in building a robust and relevant classifier (Guyon and Elisseeff 2003): First, dimension reduction is usefull to limit the risk of overfitting and increase the prediction stability of the model; second, intrepretation of the molecular signature is facilitated; third, in case of the development of diagnostic product, a restricted list is required for the subsequent validation steps (Rifai, Gillette, and Carr 2006).

Since the comprehensive analysis of all combinations of features is not computationally tractable, several selection techniques have been described, including filter (e.g., p-values thresholding), wrapper (e.g., recursive feature elimination), and embedded (e.g., sparse PLS) approaches (Saeys, Inza, and Larranaga 2007). The major challenge for such methods is to be fast and extract restricted and stable molecular signatures which still provide high performance of the classifier (Gromski et al. 2014; Determan 2015).

2 The biosigner package

The biosigner package implements a new wrapper feature selection algorithm:

  1. the dataset is split into training and testing subsets (by bootstraping, controling class proportion),

  2. model is trained on the training set and balanced accuracy is evaluated on the test set,

  3. the features are ranked according to their importance in the model,

  4. the relevant feature subset at level f is found by a binary search: a feature subset is considered relevant if and only if, when randomly permuting the intensities of other features in the test subsets, the proportion of increased or equal prediction accuracies is lower than a defined threshold f,

  5. the dataset is restricted to the selected features and steps 1 to 4 are repeated until the selected list of features is stable.

Three binary classifiers have been included in biosigner, namely PLS-DA, RF and SVM, as the performances of each machine learning approach may vary depending on the structure of the dataset (Determan 2015). The algorithm returns the tier of each feature for the selected classifer(s): tier S corresponds to the final signature, i.e., features which have been found significant in all the selection steps; features with tier A have been found significant in all but the last selection, and so on for tier B to D. Tier E regroup all previous round of selection.

As for a classical classification algorithm, the biosign method takes as input the x samples times features data frame (or matrix) of intensities, and the y factor (or character vector) of class labels (note that only binary classification is currently available). It returns the signature (signatureLs: selected feature names) and the trained model (modelLs) for each of the selected classifier. The plot method for biosign objects enable to visualize the individual boxplots of the selected features. Finally, the predict method allows to apply the trained classifier(s) on new datasets.

The algorithm has been successfully applied to transcriptomics and metabolomics data [Rinaudo et al. (2016); see also the Hands-on section below).

3 Hands-on

3.1 Loading

We first load the biosigner package:

library(biosigner)

We then use the diaplasma metabolomics dataset (Rinaudo et al. 2016) which results from the analysis of plasma samples from 69 diabetic patients were analyzed by reversed-phase liquid chromatography coupled to high-resolution mass spectrometry (LC-HRMS; Orbitrap Exactive) in the negative ionization mode. The raw data were pre-processed with XCMS and CAMERA (5,501 features), corrected for signal drift, log10 transformed, and annotated with an in-house spectral database. The patient’s age, body mass index, and diabetic type are recorded (Rinaudo et al. 2016).

data(diaplasma)

We attach diaplasma to the search path and display a summary of the content of the dataMatrix, sampleMetadata and variableMetadata with the view function from the (imported) ropls package:

attach(diaplasma)
library(ropls)
ropls::view(dataMatrix)
##         dim  class    mode typeof   size NAs min mean median max
##  69 x 5,501 matrix numeric double 3.3 Mb   0   0  4.2    4.4 8.2
##           m096.009t01.6    m096.922t00.8 ...    m995.603t10.2    m995.613t10.2
## DIA001 2.98126177377087 6.08172882312848 ... 3.93442594703862 3.96424920154706
## DIA002                0 6.13671997362279 ... 3.74201112636229 3.78128422428722
## ...                 ...              ... ...              ...              ...
## DIA077                0 6.12515971273103 ... 4.55458598372024 4.57310800324247
## DIA078 4.69123816772499   6.134420482337 ...  4.1816445335704 4.20696191303494

ropls::view(sampleMetadata, standardizeL = TRUE)
##    type     age     bmi
##  factor numeric numeric
##  nRow nCol size NAs
##    69    3 0 Mb   0
##        type age  bmi
## DIA001   T2  70 31.6
## DIA002   T2  67   28
## ...     ... ...  ...
## DIA077   T2  50   27
## DIA078   T2  65   29
## 1 data.frame 'factor' column(s) converted to 'numeric' for plotting.
## Standardization of the columns for plotting.

ropls::view(variableMetadata, standardizeL = TRUE)
##    mzmed   rtmed ... pcgroup     spiDb
##  numeric numeric ... numeric character
##   nRow nCol   size NAs
##  5,501    6 0.8 Mb   0
##                     mzmed       rtmed ... pcgroup
## m096.009t01.6 96.00899361 93.92633015 ...    1984
## m096.922t00.8 96.92192011 48.93274877 ...       4
## ...                   ...         ... ...     ...
## m995.603t10.2 995.6030195 613.4388762 ...    7160
## m995.613t10.2 995.6134422 613.4446705 ...    7161
##                                            spiDb
## m096.009t01.6 N-Acetyl-L-aspartic acid_HMDB00812
## m096.922t00.8                                   
## ...                                          ...
## m995.603t10.2                                   
## m995.613t10.2
## 3 data.frame 'character' column(s) converted to 'numeric' for plotting.
## Standardization of the columns for plotting.

We see that the diaplasma list contains three objects:

  1. dataMatrix: 69 samples x 5,501 matrix of numeric type containing the intensity profiles (log10 transformed),

  2. sampleMetadata: a 69 x 3 data frame, with the patients’

    • type: diabetic type, factor

    • age: numeric

    • bmi: body mass index, numeric

  3. variableMetadata: a 5,501 x 8 data frame, with the median m/z (‘mzmed’, numeric) and the median retention time in seconds (‘rtmed’, numeric) from XCMS, the ‘isotopes’ (character), ‘adduct’ (character) and ‘pcgroups’ (numeric) annotations from CAMERA, the names of the m/z and RT matching compounds from an in-house spectra of commercial metabolites (‘name_hmdb’, character), and the p-values resulting from the non-parametric hypothesis testing of difference in medians between types (‘type_wilcox_fdr’, numeric), and correlation with age (‘age_spearman_fdr’, numeric) and body mass index (‘bmi_spearman_fdr’, numeric), all corrected for multiple testing (False Discovery Rate).

  4. se: The previous data and metadata as a SummarizedExperiment instance

  5. eset The previous data as a ExpressionSet instance

We can observe that the 3 clinical covariates (diabetic type, age, and bmi) are stronlgy associated:

with(sampleMetadata,
plot(age, bmi, cex = 1.5, col = ifelse(type == "T1", "blue", "red"), pch = 16))
legend("topleft", cex = 1.5, legend = paste0("T", 1:2),
text.col = c("blue", "red"))

Figure 1: age, body mass index (bmi), and diabetic type of the patients from the diaplasma cohort.

3.2 Molecular signatures

Let us look for signatures of type in the diaplasma dataset by using the biosign method. To speed up computations in this demo vignette, we restrict the number of features (from 5,501 to about 500) and the number of bootstraps (5 instead of 50 [default]); the selection on the whole dataset, 50 bootstraps, and the 3 classifiers, takes around 10 min.

featureSelVl <- variableMetadata[, "mzmed"] >= 450 &
variableMetadata[, "mzmed"] < 500
sum(featureSelVl)
## [1] 533
dataMatrix <- dataMatrix[, featureSelVl]
variableMetadata <- variableMetadata[featureSelVl, ]
diaSign <- biosign(dataMatrix, sampleMetadata[, "type"], bootI = 5)
## Selecting features for the plsda model
## Selecting features for the randomforest model
## Selecting features for the svm model
## Significant features from 'S' groups:
##               plsda randomforest svm
## m495.261t08.7 "C"   "A"          "S"
## m497.284t08.1 "S"   "S"          "E"
## m497.275t08.1 "A"   "S"          "E"
## m471.241t07.6 "B"   "S"          "E"
## Accuracy:
##      plsda randomforest   svm
## Full 0.797        0.835 0.824
## AS   0.823        0.845 0.708
## S    0.825        0.858 0.708

Figure 2: Relevant signatures for the PLS-DA, Random Forest, and SVM classifiers extracted from the diaplasma dataset. The S tier corresponds to the final metabolite signature, i.e., metabolites which passed through all the selection steps.

The arguments are:

  • x: the numerical matrix (or data frame) of intensities (samples as rows, variables as columns),

  • y: the factor (or character) specifying the sample labels from the 2 classes,

  • methodVc: the classifier(s) to be used; here, the default all value means that all classifiers available (plsda, randomforest, and svm) are selected,

  • bootI: the number of bootstraps is set to 5 to speed up computations when generating this vignette; we however recommend to keep the default 50 value for your analyzes (otherwise signatures may be less stable).

  • The set.seed argument ensures that the results from this vignette can be reproduced exactly; by choosing alternative seeds (and the default bootI = 50), similar signatures are obtained, showing the stability of the selection.

Note:

  • If some features from the x matrix/data frame contain missing values (NA), these features will be removed prior to modeling with Random Forest and SVM (in contrast, the NIPALS algorithm from PLS-DA can handle missing values),

The resulting signatures for the 3 selected classifiers are both printed and plotted as tiers from S, A, up to E by decreasing relevance. The (S) tier corresponds to the final signature, i.e. features which passed through all the backward selection steps. In contrast, features from the other tiers were discarded during the last (A) or previous (B to E) selection rounds.

Note that tierMaxC = ‘A’ argument in the print and plot methods can be used to view the features from the larger S+A signatures (especially when no S features have been found, or when the performance of the S model is much lower than the S+A model).

The performance of the model built with the input dataset (balanced accuracy: mean of the sensitivity and specificity), or the subset restricted to the S or S+A signatures are shown. We see that with 1 to 5 S feature signatures (i.e., less than 1% of the input), the 3 classifiers achieve good performances (even higher than the full Random Forest and SVM models). Furthermore, reducing the number of features decreases the risk of building non-significant models (i.e., models which do not perform significantly better than those built after randomly permuting the labels). The signatures from the 3 classifiers have some distinct features, which highlights the interest of comparing various machine learning approaches.

The individual boxplots of the features from the complete signature can be visualized with:

plot(diaSign, typeC = "boxplot")

Figure 3: Individual boxplots of the features selected for at least one of the classification methods. Features selected for a single classifier are colored (red for PLS-DA, green for Random Forest and blue for SVM).

Let us see the metadata of the complete signature:

variableMetadata[getSignatureLs(diaSign)[["complete"]], ]
##                  mzmed    rtmed isotopes                        adduct pcgroup
## m495.261t08.7 495.2609 524.1249                                           1655
## m497.284t08.1 497.2840 486.5338          [M+Cl]- 462.31 [M-H]- 498.287     220
## m497.275t08.1 497.2755 486.5722          [M+Cl]- 462.31 [M-H]- 498.287     220
## m471.241t07.6 471.2408 455.5541                                          10538
##                                              spiDb
## m495.261t08.7                                     
## m497.284t08.1                                     
## m497.275t08.1 Taurochenodeoxycholic acid_HMDB00951
## m471.241t07.6

3.3 Predictions

Let us split the dataset into a training (the first 4/5th of the 183 samples) and a testing subsets, and extract the relevant features from the training subset:

trainVi <- 1:floor(0.8 * nrow(dataMatrix))
testVi <- setdiff(1:nrow(dataMatrix), trainVi)
diaTrain <- biosign(dataMatrix[trainVi, ], sampleMetadata[trainVi, "type"],
bootI = 5)
## Selecting features for the plsda model
## Selecting features for the randomforest model
## Selecting features for the svm model
## Significant features from 'S' groups:
##               plsda randomforest svm
## m497.284t08.1 "S"   "S"          "E"
## m469.215t07.8 "E"   "E"          "S"
## Accuracy:
##      plsda randomforest   svm
## Full 0.753        0.797 0.728
## AS   0.823        0.855 0.668
## S    0.814        0.782 0.603

Figure 4: Signatures from the training data set.

We extract the fitted types on the training dataset restricted to the S signatures:

diaFitDF <- predict(diaTrain)

We then print the confusion tables for each classifier:

lapply(diaFitDF, function(predFc) table(actual = sampleMetadata[trainVi,
"type"], predicted = predFc))
## $plsda
##       predicted
## actual T1 T2
##     T1 16  6
##     T2  4 29
## 
## $randomforest
##       predicted
## actual T1 T2
##     T1 14  8
##     T2  7 26
## 
## $svm
##       predicted
## actual T1 T2
##     T1  7 15
##     T2  3 30

and the corresponding balanced accuracies:

sapply(diaFitDF, function(predFc) {
conf <- table(sampleMetadata[trainVi, "type"], predFc)
conf <- sweep(conf, 1, rowSums(conf), "/")
round(mean(diag(conf)), 3)
})
##        plsda randomforest          svm 
##        0.803        0.712        0.614

Note that these values are slightly different from the accuracies returned by biosign because the latter are computed by using the resampling scheme selected by the bootI (or crossvalI) arguments:

round(getAccuracyMN(diaTrain)["S", ], 3)
##        plsda randomforest          svm 
##        0.814        0.782        0.603

Finally, we can compute the performances on the test subset:

diaTestDF <- predict(diaTrain, newdata = dataMatrix[testVi, ])
sapply(diaTestDF, function(predFc) {
conf <- table(sampleMetadata[testVi, "type"], predFc)
conf <- sweep(conf, 1, rowSums(conf), "/")
round(mean(diag(conf)), 3)
})
##        plsda randomforest          svm 
##        0.750        0.667        0.500

3.4 Working on SummarizedExperiment objects

The SummarizedExperiment class from the SummarizedExperiment bioconductor package has been developed to conveniently handle preprocessed omics objects, including the variable x sample matrix of intensities, and two DataFrames containing the sample and variable metadata, which can be accessed by the assay, colData and rowData methods respectively (remember that the data matrix is stored with samples in columns).

Getting the diaplasma dataset as a SummarizedExperiment:

# Preparing the data (matrix) and sample and variable metadata (data frames):
data(diaplasma)
data.mn <- diaplasma[["dataMatrix"]] # matrix: samples x variables
samp.df <- diaplasma[["sampleMetadata"]] # data frame: samples x sample metadata
feat.df <- diaplasma[["variableMetadata"]] # data frame: features x feature metadata

# Creating the SummarizedExperiment (package SummarizedExperiment)
library(SummarizedExperiment)
dia.se <- SummarizedExperiment(assays = list(diaplasma = t(data.mn)),
                               colData = samp.df,
                               rowData = feat.df)
# note that colData and rowData main format is DataFrame, but data frames are accepted when building the object
stopifnot(validObject(dia.se))

# Viewing the SummarizedExperiment
# ropls::view(dia.se)

The biosign method can be applied to a SummarizedExperiment object, by using the object as the x argument, and by indicating as the y argument the name of the sample metadata to be used as the response (i.e. name of the column in the colData). Note that in the example below, we restrict the data set to the first 100 features to speed up computations:

dia.se <- dia.se[1:100, ]
dia.se <- biosign(dia.se, "type", bootI = 5)

The biosign method returns the updated SummarizedExperiment object with the tiers as new columns in the rowData

feat.DF <- SummarizedExperiment::rowData(dia.se)
head(feat.DF[, grep("type_", colnames(feat.DF))])
## DataFrame with 6 rows and 3 columns
##               type_biosign_plsda type_biosign_forest type_biosign_svm
##                      <character>         <character>      <character>
## m096.009t01.6                  E                   E                B
## m096.922t00.8                  E                   E                E
## m098.025t01.3                  E                   E                E
## m099.009t01.3                  E                   B                E
## m099.009t00.9                  E                   E                E
## m099.045t04.0                  E                   A                E

and with the biosign model in the metadata slot, which can be accessed with the getBiosign method:

dia_type.biosign <- getBiosign(dia.se)
names(dia_type.biosign)
## [1] "type_plsda.forest.svm"
plot(dia_type.biosign[["type_plsda.forest.svm"]], typeC = "tier")

3.4.1 ExpressionSet format

The ExpressionSet format is currently supported as a legacy representation from the previous versions of the biosigner package (< 1.24.2) but will now be supplanted by SummarizedExperiment in future versions.

exprs, pData, and fData for ExpressionSet are similar to assay, colData and rowData for SummarizedExperiment except that assay is a list which can potentially include several matrices, and that colData and rowData are of the DataFrame format. SummarizedExperiment format further enables to store additional metadata (such as models or ggplots) in a dedicated metadata slot.

In the example below, we will first build a minimal ExpressionSet object from the diaplasma data set and view the data, and we subsequently perform the feature selection.

Getting the diaplasma data set as a ExpressionSet:

# Preparing the data (matrix) and sample and variable metadata (data frames):
data(diaplasma)
data.mn <- diaplasma[["dataMatrix"]] # matrix: samples x variables
samp.df <- diaplasma[["sampleMetadata"]] # data frame: samples x sample metadata
feat.df <- diaplasma[["variableMetadata"]] # data frame: features x feature metadata

# Creating the SummarizedExperiment (package SummarizedExperiment)
library(Biobase)
dia.eset <- Biobase::ExpressionSet(assayData = t(data.mn))
Biobase::pData(dia.eset) <- samp.df
Biobase::fData(dia.eset) <- feat.df
stopifnot(validObject(dia.eset))
# Viewing the ExpressionSet
# ropls::view(dia.eset)

Selecting the features:

dia.eset <- dia.eset[1:100, ]
dia_type.biosign <- biosign(dia.eset, "type", bootI = 5)

Note that this time, biosign returns the models an en object of the biosign class.

plot(dia_type.biosign, typeC = "tier")

The updated ExpressionSet object can be accessed with the getEset method:

dia.eset <- getEset(dia_type.biosign)
feat.df <- Biobase::fData(dia.eset)
head(feat.df[, grep("type_", colnames(feat.df))])
##               type_biosign_plsda type_biosign_forest type_biosign_svm
## m096.009t01.6                  E                   E                B
## m096.922t00.8                  E                   E                E
## m098.025t01.3                  E                   E                E
## m099.009t01.3                  E                   B                E
## m099.009t00.9                  E                   E                E
## m099.045t04.0                  E                   A                E

Before moving to new data sets, we detach diaplasma from the search path:

detach(diaplasma)

3.5 Working on MultiAssayExperiment objects

The MultiAssayExperiment format is useful to handle multi-omics data sets (???). Feature selection can be performed in parallel for each data set by applying opls to such formats. We provide an example based on the NCI60_4arrays cancer data set from the omicade4 package (which has been made available in this ropls package in the MultiAssayExperiment format).

Getting the NCI60 data set as a MultiAssayExperiment:

data("NCI60", package = "ropls")
nci.mae <- NCI60[["mae"]]
library(MultiAssayExperiment)
# Cancer types
table(nci.mae$cancer)
## 
## BR CN CO LC LE ME OV PR RE 
##  5  6  7  9  6 10  7  2  8
# Restricting to the 'ME' and 'LE' cancer types and to the 'agilent' and 'hgu95' datasets
nci.mae <- nci.mae[, nci.mae$cancer %in% c("ME", "LE"), c("agilent", "hgu95")]
## Warning: 'experiments' dropped; see 'drops()'

Performing the feature selection for each dataset:

nci.mae <- biosign(nci.mae, "cancer", bootI = 5)
## 
## 
## Selecting the features for the 'agilent' dataset:
## Selecting features for the plsda model
## Selecting features for the randomforest model
## Selecting features for the svm model
## Significant features from 'S' groups:
##          plsda randomforest svm
## VEPH1    "S"   "E"          "B"
## LHFP     "S"   "E"          "B"
## C10orf90 "B"   "E"          "S"
## EZH2     "E"   "S"          "E"
## Accuracy:
##      plsda randomforest   svm
## Full     1        1.000 1.000
## AS       1        0.900 0.983
## S        1        0.917 0.983

## 
## 
## Selecting the features for the 'hgu95' dataset:
## Selecting features for the plsda model
## Selecting features for the randomforest model
## Selecting features for the svm model
## Significant features from 'S' groups:
##         plsda randomforest svm
## TSPAN4  "S"   "S"          "E"
## TBC1D16 "S"   "E"          "B"
## NASP    "S"   "E"          "E"
## Accuracy:
##      plsda randomforest   svm
## Full     1            1 1.000
## AS       1            1 0.917
## S        1            1    NA

The biosigner method returns an updated MultiAssayExperiment with the tiers included as additional columns in the rowData of the individual SummarizedExperiment:

SummarizedExperiment::rowData(nci.mae[["agilent"]])
## DataFrame with 300 rows and 4 columns
##                  name cancer_biosign_plsda cancer_biosign_forest
##           <character>          <character>           <character>
## ST8SIA1       ST8SIA1                    E                     E
## YWHAQ           YWHAQ                    E                     E
## EPHA4           EPHA4                    E                     E
## GTPBP5         GTPBP5                    E                     E
## PVR               PVR                    E                     E
## ...               ...                  ...                   ...
## HIST1H2AB   HIST1H2AB                    E                     E
## XPO6             XPO6                    E                     E
## KIAA1688     KIAA1688                    E                     E
## TCEAL2         TCEAL2                    B                     E
## GLCCI1         GLCCI1                    E                     E
##           cancer_biosign_svm
##                  <character>
## ST8SIA1                    B
## YWHAQ                      E
## EPHA4                      B
## GTPBP5                     E
## PVR                        E
## ...                      ...
## HIST1H2AB                  E
## XPO6                       E
## KIAA1688                   E
## TCEAL2                     E
## GLCCI1                     E

The biosign model(s) are stored in the metadata of the individual SummarizedExperiment objects included in the MultiAssayExperiment, and can be accessed with the getBiosign method:

mae_biosign.ls <- getBiosign(nci.mae)
for (set.c in names(mae_biosign.ls))
plot(mae_biosign.ls[[set.c]][["cancer_plsda.forest.svm"]],
     typeC = "tier",
     plotSubC = set.c)

3.5.1 MultiDataSet objects

The MultiDataSet format (???) is currently supported as a legacy representation from the previous versions of the biosigner package (<1.24.2) but will now be supplanted by MultiAssayExperiment in future versions. Note that the mds2mae method from the MultiDataSet package enables to convert a MultiDataSet into the MultiAssayExperiment format.

Getting the NCI60 data set as a MultiDataSet:

data("NCI60", package = "ropls")
nci.mds <- NCI60[["mds"]]

Building PLS-DA models for the cancer type:

# Restricting to the "agilent" and "hgu95" datasets
nci.mds <- nci.mds[, c("agilent", "hgu95")]
# Restricting to the 'ME' and 'LE' cancer types
library(Biobase)
sample_names.vc <- Biobase::sampleNames(nci.mds[["agilent"]])
cancer_type.vc <- Biobase::pData(nci.mds[["agilent"]])[, "cancer"]
nci.mds <- nci.mds[sample_names.vc[cancer_type.vc %in% c("ME", "LE")], ]
# Selecting the features
nci_cancer.biosign <- biosign(nci.mds, "cancer", bootI = 5)

Getting back the updated MultiDataSet:

nci.mds <- getMset(nci_cancer.biosign)

4 Extraction of biomarker signatures from other omics datasets

In this section, biosign is applied to two metabolomics and one transcriptomics data sets. Please refer to Rinaudo et al. (2016) for a full discussion of the methods and results.

4.1 Physiological variations of the human urine metabolome (metabolomics)

The sacurine LC-HRMS dataset from the dependent ropls package can also be used (Thevenot et al. 2015): Urine samples from a cohort of 183 adults 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. 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 (see the ropls vignette for further details and examples).

We can for instance look for signatures of the gender:

data(sacurine)
sacSign <- biosign(sacurine[["dataMatrix"]],
sacurine[["sampleMetadata"]][, "gender"],
methodVc = "plsda")
## Selecting features for the plsda model
## Significant features from 'S' groups:
##                          plsda
## Malic acid               "S"  
## p-Anisic acid            "S"  
## Testosterone glucuronide "S"  
## Accuracy:
##      plsda
## Full 0.876
## AS   0.882
## S    0.889

Figure 5: PLS-DA signature from the ‘sacurine’ data set.

4.2 Apples spikes with known compounds (metabolomics)

The spikedApples dataset was obtained by LC-HRMS analysis (SYNAPT Q-TOF, Waters) of one control and three spiked groups of 10 apples each. The spiked mixtures consists in 2 compounds which were not naturally present in the matrix and 7 compounds aimed at achieving a final increase of 20%, 40% or 100% of the endogeneous concentrations. The authors identified 22 features (out of the 1,632 detected in the positive ionization mode; i.e. 1.3%) which came from the spiked compounds. The dataset is included in the BioMark R package (Franceschi et al. 2012). Let us use the control and group1 samples (20 in total) in this study.

library(BioMark)
data(SpikePos)
group1Vi <- which(SpikePos[["classes"]] %in% c("control", "group1"))
appleMN <- SpikePos[["data"]][group1Vi, ]
spikeFc <- factor(SpikePos[["classes"]][group1Vi])
annotDF <- SpikePos[["annotation"]]
rownames(annotDF) <- colnames(appleMN)

We can check, by using the opls method from the ropls package for multivariate analysis, that:

  1. no clear separation can be observed by PCA:
biomark.pca <- ropls::opls(appleMN, fig.pdfC = "none")
## PCA
## 20 samples x 1632 variables
## standard scaling of predictors
##       R2X(cum) pre ort
## Total    0.523   7   0
ropls::plot(biomark.pca, parAsColFcVn = spikeFc)

  1. PLS-DA modeling with the full dataset is not significant (as seen on the top left plot: 7 out of 20 models trained after random permutations of the labels have Q2 values higher than the model trained with the true labels):
biomark.pls <- ropls::opls(appleMN, spikeFc)
## PLS-DA
## 20 samples x 1632 variables and 1 response
## standard scaling of predictors and response(s)
##       R2X(cum) R2Y(cum) Q2(cum)  RMSEE pre ort pR2Y  pQ2
## Total    0.145    0.995     0.4 0.0396   2   0 0.05 0.55

Let us now extract the molecular signatures:

appleSign <- biosign(appleMN, spikeFc)
## Selecting features for the plsda model
## Selecting features for the randomforest model
## Selecting features for the svm model
## Significant features from 'S' groups:
##           plsda randomforest svm
## 449.1/327 "S"   "S"          "C"
## Accuracy:
##      plsda randomforest   svm
## Full  0.79        0.921 0.793
## AS    1.00        1.000 0.853
## S     1.00        1.000    NA

The 449.1/327 corresponds to the Cyanidin-3-galactoside (absent in the control; Franceschi et al. (2012)).

annotDF <- SpikePos[["annotation"]]
rownames(annotDF) <- colnames(appleMN)
annotDF[getSignatureLs(appleSign)[["complete"]], c("adduct", "found.in.standards")]
##           adduct found.in.standards
## 449.1/327                         1

4.3 Bone marrow from acute leukemia patients (transcriptomics)

Samples from 47 patients with acute lymphoblastic leukemia (ALL) and 25 patients with acute myeloid leukemia (AML) have been analyzed using Affymetrix Hgu6800 chips resulting in expression data of 7,129 gene probes (Golub et al. 1999). The golub dataset is available in the golubEsets package from Bioconductor in the ExpressionSet format. Let us compute for example the SVM signature (to speed up this demo example, the number of features is restricted to 500):

library(golubEsets)
data(Golub_Merge)
Golub_Merge
## ExpressionSet (storageMode: lockedEnvironment)
## assayData: 7129 features, 72 samples 
##   element names: exprs 
## protocolData: none
## phenoData
##   sampleNames: 39 40 ... 33 (72 total)
##   varLabels: Samples ALL.AML ... Source (11 total)
##   varMetadata: labelDescription
## featureData: none
## experimentData: use 'experimentData(object)'
##   pubMedIds: 10521349 
## Annotation: hu6800
# restricting to the last 500 features
golub.eset <- Golub_Merge[1501:2000, ]
table(Biobase::pData(golub.eset)[, "ALL.AML"])
## 
## ALL AML 
##  47  25
golubSign <- biosign(golub.eset, "ALL.AML", methodVc = "svm")
## Selecting features for the svm model
## Significant features from 'S' groups:
##           svm
## M11147_at "S"
## M17733_at "S"
## M19507_at "S"
## M27891_at "S"
## Accuracy:
##        svm
## Full 0.955
## AS   0.967
## S    0.959

Figure 6: SVM signature from the golub data set.

The computation results in a signature of 4 features only and a sparse SVM model performing even better (95.9% accuracy) than the model trained on the dataset of 500 variables (95.5% accuracy).

The hu6800.db bioconductor package can be used to get the annotation of the selected probes (Carlson 2016):

library(hu6800.db)
sapply(getSignatureLs(golubSign)[["complete"]],
       function(probeC)
       get(probeC, env = hu6800GENENAME))
##                  M11147_at                  M17733_at 
##     "ferritin light chain" "thymosin beta 4 X-linked" 
##                  M19507_at                  M27891_at 
##          "myeloperoxidase"               "cystatin C"

Cystatin C is part of the 50 gene signature selected by Golub and colleagues on the basis of a metric derived from the Student’s statistic of mean differences between the AML and ALL groups (Golub et al. 1999). Interestingly, the third probe, myeloperoxidase, is a cytochemical marker for the diagnosis (and also potentially the prognosis) of acute myeloid leukemia (AML).

5 Session info

Here is the output of sessionInfo() on the system on which this document was compiled:

## R version 4.3.1 (2023-06-16)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Ubuntu 22.04.3 LTS
## 
## Matrix products: default
## BLAS:   /home/biocbuild/bbs-3.18-bioc/R/lib/libRblas.so 
## LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0
## 
## locale:
##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
##  [3] LC_TIME=en_GB              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       
## 
## time zone: America/New_York
## tzcode source: system (glibc)
## 
## attached base packages:
## [1] stats4    stats     graphics  grDevices utils     datasets  methods  
## [8] base     
## 
## other attached packages:
##  [1] hu6800.db_3.13.0            org.Hs.eg.db_3.18.0        
##  [3] AnnotationDbi_1.64.0        golubEsets_1.43.0          
##  [5] BioMark_0.4.5               st_1.2.7                   
##  [7] sda_1.3.8                   fdrtool_1.2.17             
##  [9] corpcor_1.6.10              entropy_1.3.1              
## [11] MASS_7.3-60                 glmnet_4.1-8               
## [13] Matrix_1.6-1.1              pls_2.8-2                  
## [15] MultiAssayExperiment_1.28.0 SummarizedExperiment_1.32.0
## [17] Biobase_2.62.0              GenomicRanges_1.54.0       
## [19] GenomeInfoDb_1.38.0         IRanges_2.36.0             
## [21] S4Vectors_0.40.0            BiocGenerics_0.48.0        
## [23] MatrixGenerics_1.14.0       matrixStats_1.0.0          
## [25] ropls_1.34.0                biosigner_1.30.0           
## [27] BiocStyle_2.30.0           
## 
## loaded via a namespace (and not attached):
##  [1] blob_1.2.4              Biostrings_2.70.0       bitops_1.0-7           
##  [4] fastmap_1.1.1           RCurl_1.98-1.12         digest_0.6.33          
##  [7] survival_3.5-7          statmod_1.5.0           KEGGREST_1.42.0        
## [10] RSQLite_2.3.1           magrittr_2.0.3          compiler_4.3.1         
## [13] rlang_1.1.1             sass_0.4.7              tools_4.3.1            
## [16] yaml_2.3.7              calibrate_1.7.7         knitr_1.44             
## [19] S4Arrays_1.2.0          bit_4.0.5               DelayedArray_0.28.0    
## [22] abind_1.4-5             grid_4.3.1              e1071_1.7-13           
## [25] iterators_1.0.14        cli_3.6.1               rmarkdown_2.25         
## [28] crayon_1.5.2            httr_1.4.7              BiocBaseUtils_1.4.0    
## [31] DBI_1.1.3               cachem_1.0.8            proxy_0.4-27           
## [34] zlibbioc_1.48.0         splines_4.3.1           BiocManager_1.30.22    
## [37] XVector_0.42.0          vctrs_0.6.4             jsonlite_1.8.7         
## [40] bookdown_0.36           bit64_4.0.5             qqman_0.1.9            
## [43] magick_2.8.1            foreach_1.5.2           limma_3.58.0           
## [46] jquerylib_0.1.4         MultiDataSet_1.30.0     codetools_0.2-19       
## [49] shape_1.4.6             htmltools_0.5.6.1       randomForest_4.7-1.1   
## [52] GenomeInfoDbData_1.2.11 R6_2.5.1                evaluate_0.22          
## [55] lattice_0.22-5          png_0.1-8               memoise_2.0.1          
## [58] bslib_0.5.1             class_7.3-22            Rcpp_1.0.11            
## [61] SparseArray_1.2.0       xfun_0.40               pkgconfig_2.0.3

References

Carlson, M. 2016. Hu6800.db: Affymetrix Hugenefl Genome Array Annotation Data (Chip Hu6800).

Determan, CE. 2015. “Optimal Algorithm for Metabolomics Classification and Feature Selection Varies by Dataset.” International Journal of Biology 7 (1): 100–115. https://doi.org/10.5539/ijb.v7n1p100.

Franceschi, P., D. Masuero, U. Vrhovsek, F. Mattivi, and R. Wehrens. 2012. “A Benchmark Spike-in Data Set for Biomarker Identification in Metabolomics.” Journal of Chemometrics 26 (1-2): 16–24. https://doi.org/10.1002/cem.1420.

Golub, TR., DK. Slonim, P. Tamayo, C. Huard, M. Gaasenbeek, JP. Mesirov, H. Coller, et al. 1999. “Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring.” Science 286 (5439): 531. https://doi.org/10.1126/science.286.5439.531.

Gromski, PS., Y. Xu, E. Correa, DI. Ellis, ML. Turner, and R. Goodacre. 2014. “A Comparative Investigation of Modern Feature Selection and Classification Approaches for the Analysis of Mass Spectrometry Data.” Analytica Chimica Acta 829 (0): 1–8. https://doi.org/10.1016/j.aca.2014.03.039.

Guyon, I., and A. Elisseeff. 2003. “An Introduction to Variable and Feature Selection.” Journal of Machine Learning Research 3: 1157–82.

Rifai, N., MA. Gillette, and SA. Carr. 2006. “Protein Biomarker Discovery and Validation: The Long and Uncertain Path to Clinical Utility.” Nature Biotechnology. https://doi.org/10.1038/nbt1235.

Rinaudo, P., S. Boudah, C. Junot, and EA. Thevenot. 2016. “Biosigner: A New Method for the Discovery of Significant Molecular Signatures from Omics Data.” Frontiers in Molecular Biosciences 3. https://doi.org/10.3389/fmolb.2016.00026.

Saeys, Y., I. Inza, and P. Larranaga. 2007. “A Review of Feature Selection Techniques in Bioinformatics.” Bioinformatics 23 (19): 2507–17. https://doi.org/10.1093/bioinformatics/btm344.

Thevenot, EA., A. Roux, Y. Xu, 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. https://doi.org/10.1021/acs.jproteome.5b00354.