Returns the predictions in an 'MSnSet'

Convenience accessor to the predicted feature localisation in an 'MSnSet'. This function returns the predictions of an MSnSet instance. As a side effect, it prints out a prediction table.

getPredictions(object, fcol, scol, mcol = "markers", t = 0, verbose = TRUE)

Arguments

object: An instance of class "MSnSet".
fcol: The name of the prediction column in the featureData slot.
scol: The name of the prediction score column in the featureData slot. If missing, created by pasting '.scores' after fcol.
mcol: The feature meta data column containing the labelled training data.
t: The score threshold. Predictions with score < t are set to 'unknown'. Default is 0. It is also possible to define thresholds for each prediction class, in which case, t is a named numeric with names exactly matching the unique prediction class names.
verbose: If TRUE, a prediction table is printed and the predictions are returned invisibly. If FALSE, the predictions are returned.

Value

An instance of class "MSnSet" with fcol.pred feature variable storing the prediction results according to the chosen threshold.

Author

Laurent Gatto and Lisa Breckels

Examples

library("pRolocdata")
data(dunkley2006)
res <- svmClassification(dunkley2006, fcol = "pd.markers",
                         sigma = 0.1, cost = 0.5)
#> [1] "pd.markers"
fData(res)$svm[500:510]
#>  [1] Plastid     Plastid     ER membrane Ribosome    Ribosome    Ribosome   
#>  [7] Ribosome    Ribosome    Ribosome    Ribosome    Ribosome   
#> 9 Levels: ER lumen ER membrane Golgi Mitochondrion PM Plastid Ribosome ... vacuole
fData(res)$svm.scores[500:510]
#>  [1] 0.6593303 0.7701427 0.6752305 0.4882591 0.5751725 0.5766412 0.6376513
#>  [8] 0.6215169 0.5313678 0.6123313 0.6602089
getPredictions(res, fcol = "svm", t = 0) ## all predictions
#> ans
#>      ER lumen   ER membrane         Golgi Mitochondrion            PM 
#>            16           188           101           101           125 
#>       Plastid      Ribosome           TGN       vacuole 
#>            52            59            17            30 
#> MSnSet (storageMode: lockedEnvironment)
#> assayData: 689 features, 16 samples 
#>   element names: exprs 
#> protocolData: none
#> phenoData
#>   sampleNames: M1F1A M1F4A ... M2F11B (16 total)
#>   varLabels: membrane.prep fraction replicate
#>   varMetadata: labelDescription
#> featureData
#>   featureNames: AT1G09210 AT1G21750 ... AT4G39080 (689 total)
#>   fvarLabels: assigned evidence ... svm.pred (11 total)
#>   fvarMetadata: labelDescription
#> experimentData: use 'experimentData(object)'
#>   pubMedIds: 16618929 
#> Annotation:  
#> - - - Processing information - - -
#> Loaded on Thu Jul 16 22:53:08 2015. 
#> Normalised to sum of intensities. 
#> Added markers from  'mrk' marker vector. Thu Jul 16 22:53:08 2015 
#> Performed svm prediction (cost=0.5 sigma=0.1) Tue Mar 12 07:13:29 2024 
#> Added svm predictions according to global threshold = 0 Tue Mar 12 07:13:29 2024 
#>  MSnbase version: 1.17.12 
getPredictions(res, fcol = "svm", t = .9) ## single threshold 
#> ans
#>      ER lumen   ER membrane         Golgi Mitochondrion            PM 
#>            14            56            28            55            46 
#>       Plastid      Ribosome           TGN       unknown       vacuole 
#>            20            19            13           417            21 
#> MSnSet (storageMode: lockedEnvironment)
#> assayData: 689 features, 16 samples 
#>   element names: exprs 
#> protocolData: none
#> phenoData
#>   sampleNames: M1F1A M1F4A ... M2F11B (16 total)
#>   varLabels: membrane.prep fraction replicate
#>   varMetadata: labelDescription
#> featureData
#>   featureNames: AT1G09210 AT1G21750 ... AT4G39080 (689 total)
#>   fvarLabels: assigned evidence ... svm.pred (11 total)
#>   fvarMetadata: labelDescription
#> experimentData: use 'experimentData(object)'
#>   pubMedIds: 16618929 
#> Annotation:  
#> - - - Processing information - - -
#> Loaded on Thu Jul 16 22:53:08 2015. 
#> Normalised to sum of intensities. 
#> Added markers from  'mrk' marker vector. Thu Jul 16 22:53:08 2015 
#> Performed svm prediction (cost=0.5 sigma=0.1) Tue Mar 12 07:13:29 2024 
#> Added svm predictions according to global threshold = 0.9 Tue Mar 12 07:13:29 2024 
#>  MSnbase version: 1.17.12 
## 50% top predictions per class
ts <- orgQuants(res, fcol = "svm", t = .5)
#>      ER lumen   ER membrane         Golgi Mitochondrion            PM 
#>     0.2995766     0.8368847     0.7805362     0.7484314     0.7302249 
#>       Plastid      Ribosome           TGN       vacuole 
#>     0.7746137     0.5428105     0.5276547     0.5704931 
getPredictions(res, fcol = "svm", t = ts)
#> ans
#>      ER lumen   ER membrane         Golgi Mitochondrion            PM 
#>            15           117            65            78            86 
#>       Plastid      Ribosome           TGN       unknown       vacuole 
#>            36            39            15           212            26 
#> MSnSet (storageMode: lockedEnvironment)
#> assayData: 689 features, 16 samples 
#>   element names: exprs 
#> protocolData: none
#> phenoData
#>   sampleNames: M1F1A M1F4A ... M2F11B (16 total)
#>   varLabels: membrane.prep fraction replicate
#>   varMetadata: labelDescription
#> featureData
#>   featureNames: AT1G09210 AT1G21750 ... AT4G39080 (689 total)
#>   fvarLabels: assigned evidence ... svm.pred (11 total)
#>   fvarMetadata: labelDescription
#> experimentData: use 'experimentData(object)'
#>   pubMedIds: 16618929 
#> Annotation:  
#> - - - Processing information - - -
#> Loaded on Thu Jul 16 22:53:08 2015. 
#> Normalised to sum of intensities. 
#> Added markers from  'mrk' marker vector. Thu Jul 16 22:53:08 2015 
#> Performed svm prediction (cost=0.5 sigma=0.1) Tue Mar 12 07:13:29 2024 
#> Added svm predictions according to thresholds: ER lumen = 0.30, ER membrane = 0.84, Golgi = 0.78, Mitochondrion = 0.75, PM = 0.73, Plastid = 0.77, Ribosome = 0.54, TGN = 0.53, vacuole = 0.57 Tue Mar 12 07:13:29 2024 
#>  MSnbase version: 1.17.12

Arguments

Value

See also

Author

Examples