step_opls_denoise creates a 'specification' of a recipe
step that will filter the first orthogonal component of the OPLS
transfomation on the columns.
step_opls_denoise( recipe, ..., role = NA, trained = FALSE, outcome = NULL, Wortho = NULL, Portho = NULL, skip = FALSE, id = rand_id("opls_denoise") ) # S3 method for step_opls_denoise tidy(x, ...)
| recipe | A recipe object. The step will be added to the sequence of operations for this recipe. |
|---|---|
| ... | One or more selector functions to choose which
variables are affected by the step. See |
| role | Not used by this step since no new variables are created. |
| trained | A logical to indicate if the quantities for preprocessing have been estimated. |
| outcome | When a single outcome is available, character
string or call to |
| Wortho | A vector a weights for the first orthogonal component. This is
|
| Portho | A vector of loadings for the first orthogonal component. This is
|
| skip | A logical. Should the step be skipped when the
recipe is baked by |
| id | A character string that is unique to this step to identify it. |
| x | A |
An updated version of recipe with the new step
added to the sequence of existing steps (if any). For the
tidy method, a tibble with columns terms (the
selectors or variables selected), value (the
standard deviations and means), and statistic for the type of value.
Orthogonal Projection to Latent Structurees (OPLS) allows the separation of the predictor variations that are correlated and orthogonal to the response. This allows to remove systematic variation that are not correlated to the response.
The OPLS algorithm is implemented only for binary outcomes!
OPLS calculation uses the implementation of the R package: https://bioconductor.org/packages/release/bioc/html/ropls.html
Trygg, J., & Wold, S. (2002). Orthogonal projections to latent structures (O-PLS). Journal of Chemometrics, 16(3), 119–128. doi:10.1002/cem.695 https://onlinelibrary.wiley.com/doi/abs/10.1002/cem.695
Thévenot, E. A., Roux, A., Xu, Y., Ezan, E., & Junot, C. (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), 3322–3335. doi:10.1021/acs.jproteome.5b00354 https://pubs.acs.org/doi/10.1021/acs.jproteome.5b00354
#>#> ✔ broom 0.7.6 ✔ purrr 0.3.4 #> ✔ dials 0.0.9 ✔ recipes 0.1.15 #> ✔ dplyr 1.0.5 ✔ rsample 0.0.9 #> ✔ ggplot2 3.3.3 ✔ tidyr 1.1.3 #> ✔ infer 0.5.4 ✔ tune 0.1.3 #> ✔ modeldata 0.1.0 ✔ workflows 0.2.2 #> ✔ parsnip 0.1.5 ✔ yardstick 0.0.8#> Conflicts ───────────────────────────────────────── tidymodels_conflicts() ── #> ✖ purrr::discard() masks scales::discard() #> ✖ tidyr::extract() masks tidySpectR::extract() #> ✖ dplyr::filter() masks stats::filter() #> ✖ dplyr::lag() masks stats::lag() #> ✖ recipes::step() masks stats::step()library(tidySpectR) data(sacurine) attach(sacurine) genderFc <- sampleMetadata[, "gender"] urinedata <- dataMatrix %>% cbind(genderFc) %>% as_tibble() %>% add_column(id = rownames(dataMatrix), .before = 1) %>% select(-id) rec <- recipe(urinedata, genderFc ~.) %>% step_normalize(all_predictors()) %>% step_opls_denoise(all_predictors(), outcome = "genderFc") tidy(rec)#> # A tibble: 2 x 6 #> number operation type trained skip id #> <int> <chr> <chr> <lgl> <lgl> <chr> #> 1 1 step normalize FALSE FALSE normalize_1OPbL #> 2 2 step opls_denoise FALSE FALSE opls_denoise_547042rec %>% prep() %>% bake(NULL)#> # A tibble: 183 x 110 #> genderFc `(2-methoxyethox… `(gamma)Glu-Leu… `1-Methyluric a… `1-Methylxanthi… #> <dbl> <dbl> <dbl> <dbl> <dbl> #> 1 1 -0.770 -1.20 -0.102 -0.378 #> 2 2 0.160 0.0410 -0.795 -0.762 #> 3 1 -0.180 -0.104 0.273 0.780 #> 4 1 -1.37 0.147 1.05 0.723 #> 5 1 0.0986 1.27 -0.210 -0.263 #> 6 1 0.555 0.328 0.687 0.774 #> 7 1 -0.572 -0.224 0.438 0.635 #> 8 1 -0.384 1.38 -0.478 -0.584 #> 9 2 -0.521 -1.89 0.272 -0.244 #> 10 1 0.140 -1.51 1.02 1.14 #> # … with 173 more rows, and 105 more variables: 1,3-Dimethyluric acid <dbl>, #> # 1,7-Dimethyluric acid <dbl>, 2-acetamido-4-methylphenyl acetate <dbl>, #> # 2-Aminoadipic acid <dbl>, 2-Hydroxybenzyl alcohol <dbl>, #> # 2-Isopropylmalic acid <dbl>, 2-Methylhippuric acid <dbl>, #> # 2,2-Dimethylglutaric acid <dbl>, 3-Hydroxybenzyl alcohol <dbl>, #> # 3-Hydroxyphenylacetic acid <dbl>, #> # 3-Indole carboxylic acid glucuronide <dbl>, #> # 3-Methyl-2-oxovaleric acid <dbl>, 3-Methylcrotonylglycine <dbl>, #> # 3,3-Dimethylglutaric acid <dbl>, 3,4-Dihydroxybenzeneacetic acid <dbl>, #> # 3,5-dihydroxybenzoic acid/3,4-dihydroxybenzoic acid <dbl>, #> # 3,7-Dimethyluric acid <dbl>, 4-Acetamidobutanoic acid isomer 2 <dbl>, #> # 4-Acetamidobutanoic acid isomer 3 <dbl>, 4-Hydroxybenzoic acid <dbl>, #> # 4-Methylhippuric acid/3-Methylhippuric acid <dbl>, #> # 5-Hydroxyindoleacetic acid <dbl>, 5-Sulfosalicylic acid <dbl>, #> # 6-(2-hydroxyethoxy)-6-oxohexanoic acid <dbl>, #> # 6-(carboxymethoxy)-hexanoic acid <dbl>, 9-Methylxanthine <dbl>, #> # Acetaminophen glucuronide <dbl>, Acetylphenylalanine <dbl>, #> # alpha-N-Phenylacetyl-glutamine <dbl>, Aminosalicyluric acid <dbl>, #> # Asp-Leu/Ile isomer 1 <dbl>, Asp-Leu/Ile isomer 2 <dbl>, #> # Aspartic acid <dbl>, Azelaic acid <dbl>, Benzoic acid isomer <dbl>, #> # Chenodeoxycholic acid isomer <dbl>, Cinnamoylglycine <dbl>, #> # Citric acid <dbl>, Dehydroepiandrosterone 3-glucuronide <dbl>, #> # Dehydroepiandrosterone sulfate <dbl>, Deoxyhexose <dbl>, #> # Dimethylguanosine <dbl>, FMNH2 <dbl>, Fumaric acid <dbl>, #> # Gentisic acid <dbl>, Glu-Val <dbl>, Gluconic acid and/or isomers <dbl>, #> # Glucuronic acid and/or isomers <dbl>, Glyceric acid <dbl>, #> # Glycocholic acid isomer 2 <dbl>, Glycocholic acid isomer 3 <dbl>, #> # Heptylmalonic acid <dbl>, Hexanoylglycine <dbl>, Hippuric acid <dbl>, #> # Hydroxybenzyl alcohol isomer <dbl>, Hydroxyphenyllactic acid <dbl>, #> # Hydroxysuberic acid isomer 1 <dbl>, Hydroxysuberic acid isomer 2 <dbl>, #> # Isovalerylalanine isomer <dbl>, Ketoleucine <dbl>, Kynurenic acid <dbl>, #> # Malic acid <dbl>, Methoxysalicylic acid isomer <dbl>, #> # Methyl (hydroxymethyl)pyrrolidine-carboxylate/Methyl (hydroxy)piperidine-carboxylate <dbl>, #> # Methylinosine <dbl>, Mevalonic acid isomer 1 <dbl>, #> # Monoethyl phthalate <dbl>, N-Acetyl-aspartic acid <dbl>, #> # N-Acetylisoleucine <dbl>, N-Acetylleucine <dbl>, N-Acetyltryptophan <dbl>, #> # N-Acetyltryptophan isomer 3 <dbl>, N2-Acetylaminoadipic acid <dbl>, #> # N4-Acetylcytidine <dbl>, Nicotinuric acid isomer <dbl>, #> # Ortho-Hydroxyphenylacetic acid <dbl>, Oxoglutaric acid <dbl>, #> # p-Anisic acid <dbl>, p-Hydroxyhippuric acid <dbl>, #> # p-Hydroxymandelic acid <dbl>, p-Hydroxyphenylacetic acid <dbl>, #> # Pantothenic acid <dbl>, Pentose <dbl>, Phe-Tyr-Asp (and isomers) <dbl>, #> # Porphobilinogen <dbl>, Pyridoxic acid isomer 1 <dbl>, #> # Pyridylacetylglycine <dbl>, Pyrocatechol sulfate <dbl>, #> # Pyroglutamic acid <dbl>, Pyrroledicarboxylic acid <dbl>, #> # Pyruvic acid <dbl>, Quinic acid <dbl>, Salicylic acid <dbl>, #> # Sebacic acid <dbl>, Suberic acid <dbl>, Sulfosalicylic acid isomer <dbl>, #> # Taurine <dbl>, Testosterone glucuronide <dbl>, #> # Tetrahydrohippuric acid <dbl>, Threo-3-Phenylserine <dbl>, …