Hello! I am having four errors in ROC Curve Analysis in MetaboAnalyst R 4.0 when using the example data. Three of them were reported as could not find the function. Each error was labeled in bold and italic.
Part 1. Classical ROC Curve Analysis
mSet<-SetAnalysisMode(mSet, “univ”)
mSet<-PrepareROCData(mSet)
OPTION 1 Perform univariate ROC curve analysis
mSet<-Perform.UnivROC(mSet, feat.nm = “Isoleucine”, version = “Isoleucine”, “png”, dpi=300, isAUC=F, isOpt=T, optMethod=“closest.topleft”, isPartial=F, measure=“sp”, cutoff=0.2)
mSet<-PlotRocUnivBoxPlot(mSet, “Isoleucine”, “Isoleucineboxplot_0_”, “png”, 72, T, FALSE)
#Error 1 >could not find function “PlotBoxPlot”
#Note: For this error, I change the function to PlotRocUnivBoxPlot(), then it seems to work but I am not sure if it’s right to do so.
mSet<-CalculateFeatureRanking(mSet)
OPTION 2 Perform univariate ROC curve analysis, resulting in a partial AUC with a 95% CI band
#This line runs well with no error
mSet<-Perform.UnivROC(mSet, feat.nm = “Valine”, version = “Valine”, “png”, dpi=300,
isAUC=T, isOpt=T, optMethod=“closest.topleft”, isPartial=T, measure=“se”, cutoff=0.2)
OPTION 3 Perform univariate ROC curve analysis on a metabolite ratio pair
mSet<-Perform.UnivROC(mSet, feat.nm = “Isoleucine/Valine”, version = “IsoleucineValine”,
“png”, dpi=300, isAUC=T, isOpt=T, optMethod=“closest.topleft”,
isPartial=T, measure=“se”, cutoff=0.2)
#Error 2 >Error in[.data.frame(data_ori_norm, , feat.nm) : Undefined columns are selected
Part 2. Multivariate ROC Curve Explorer
mSet<-SetAnalysisMode(mSet, “explore”)
mSet<-PrepareROCData(mSet)
mSet<-PerformCV.explore(mSet, cls.method = “svm”, rank.method = “svm”, lvNum = 2)
#>Warning messages:
1: In svm.default(x.in, y.in, type = “C”, kernel = “linear”) :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
2: In svm.default(x.train, y.train, type = “C”, kernel = “linear”, :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
3: In svm.default(x.in, y.in, type = “C”, kernel = “linear”) :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
4: In svm.default(x.train, y.train, type = “C”, kernel = “linear”, :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
5: In svm.default(x.in, y.in, type = “C”, kernel = “linear”) :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
6: In svm.default(x.train, y.train, type = “C”, kernel = “linear”, :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
7: In svm.default(x.in, y.in, type = “C”, kernel = “linear”) :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
8: In svm.default(x.train, y.train, type = “C”, kernel = “linear”, :
Variable(s) ‘Creatinine’ constant. Cannot scale data.
OPTION 1 Comparison plot of ROC curves of all models
mSet<-PlotROC(mSet, imgName = “ROC_all_models”, format = “png”, dpi = 300, mdl.inx= 0, avg.method = “threshold”, show.conf = 0, show.holdout = 0, focus=“fpr”, cutoff=0.5)
mSet<-PlotProbView(mSet, imgName = “multi_roc_prob”, format = “png”, dpi = 300,
mdl.inx = -1, show = 0, showPred = 0)
mSet<-PlotAccuracy(mSet, imgName = “multi_roc_accuracy”, format = “png”, dpi = 300)
mSet<-PlotImpVars(mSet, imgName = “multi_roc_impvar”, format=“png”, dpi=300,
mdl.inx = -1, measure=“freq”, feat.num=15)
#Error 3 >Error in PlotImpVars(mSet, imgName = “multi_roc_impvar”, format = “png”, : could not find function “PlotImpVars”
OPTION 2 Plot the ROC curve of a single selected model, in this case model 1 and display the confidence interval
mSet<-PlotROC(mSet, imgName = “ROC_model1”, format = “png”, dpi = 300,
mdl.inx = 1, avg.method = “threshold”, show.conf = 1, 0, “fpr”, 0.2)
imp.feats<-GetImpFeatureMat(mSet, mSet$analSet$multiROC$imp.cv, 10)
head(imp.feats)
#Error 4 >Error in GetImpFeatureMat(mSet, mSet$analSet$multiROC$imp.cv, 10) :
could not find function “GetImpFeatureMat”
My session information:
R version 4.2.1 (2022-06-23 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 22631)
Matrix products: default
locale:
[1] LC_COLLATE=Chinese (Simplified)_China.utf8 LC_CTYPE=Chinese (Simplified)_China.utf8
[3] LC_MONETARY=Chinese (Simplified)_China.utf8 LC_NUMERIC=C
[5] LC_TIME=Chinese (Simplified)_China.utf8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] Rserve_1.8-13 MetaboAnalystR_4.0.0
loaded via a namespace (and not attached):
[1] fgsea_1.22.0 colorspace_2.1-0 class_7.3-22 siggenes_1.70.0
[5] proxy_0.4-27 rstudioapi_0.16.0 listenv_0.9.1 farver_2.1.2
[9] bit64_4.0.5 prodlim_2024.06.25 fansi_1.0.6 lubridate_1.9.3
[13] codetools_0.2-20 splines_4.2.0 cachem_1.1.0 impute_1.70.0
[17] scrime_1.3.5 glasso_1.11 jsonlite_1.8.8 pROC_1.18.5
[21] Cairo_1.6-2 caret_6.0-94 pheatmap_1.0.12 compiler_4.2.0
[25] httr_1.4.7 Matrix_1.5-3 fastmap_1.2.0 lazyeval_0.2.2
[29] limma_3.52.4 cli_3.6.3 crmn_0.0.21 htmltools_0.5.8.1
[33] tools_4.2.0 igraph_2.0.3 gtable_0.3.5 glue_1.7.0
[37] reshape2_1.4.4 dplyr_1.1.4 fastmatch_1.1-4 Rcpp_1.0.12
[41] Biobase_2.56.0 vctrs_0.6.5 multtest_2.52.0 preprocessCore_1.58.0
[45] nlme_3.1-165 iterators_1.0.14 timeDate_4032.109 gower_1.0.1
[49] stringr_1.5.1 globals_0.16.3 timechange_0.3.0 lifecycle_1.0.4
[53] gtools_3.9.5 future_1.33.2 edgeR_3.38.4 MASS_7.3-56
[57] scales_1.3.0 ipred_0.9-14 pcaMethods_1.88.0 parallel_4.2.0
[61] RColorBrewer_1.1-3 qs_0.26.3 memoise_2.0.1 gridExtra_2.3
[65] ggplot2_3.5.1 rpart_4.1.23 stringi_1.8.4 RSQLite_2.3.7
[69] foreach_1.5.2 e1071_1.7-14 caTools_1.18.2 BiocGenerics_0.42.0
[73] hardhat_1.4.0 BiocParallel_1.30.4 lava_1.8.0 rlang_1.1.4
[77] pkgconfig_2.0.3 bitops_1.0-7 lattice_0.22-6 ROCR_1.0-11
[81] purrr_1.0.2 recipes_1.1.0 htmlwidgets_1.6.4 labeling_0.4.3
[85] cowplot_1.1.3 bit_4.0.5 tidyselect_1.2.1 parallelly_1.37.1
[89] plyr_1.8.9 magrittr_2.0.3 R6_2.5.1 gplots_3.1.3.1
[93] generics_0.1.3 DBI_1.2.3 pillar_1.9.0 withr_3.0.0
[97] survival_3.7-0 nnet_7.3-19 tibble_3.2.1 future.apply_1.11.2
[101] KernSmooth_2.23-24 utf8_1.2.4 RApiSerialize_0.1.3 plotly_4.10.4
[105] locfit_1.5-9.10 grid_4.2.0 data.table_1.15.4 blob_1.2.4
[109] ModelMetrics_1.2.2.2 digest_0.6.36 xtable_1.8-4 tidyr_1.3.1
[113] RcppParallel_5.1.8 stats4_4.2.0 munsell_0.5.1 stringfish_0.16.0
[117] viridisLite_0.4.2 sessioninfo_1.2.2
Thank you for your guidance!