Skip to contents

Predict clinical and molecular features

Source: R/estimate.features.R
estimate.features.Rd

Main function used to predict various clinical and molecular features from gene-level methylation data in prostate cancer patients.

Usage

estimate.features(
  gene.methy.data,
  models,
  prop.missing.cutoff = 0.3,
  validate.data = TRUE
)

Arguments

gene.methy.data

A data frame with gene-level methylation data, created by gene.methylation. Patients are rows and columns are genes.

models

A list of models used to predict features from gene-level methylation data. The models should come from data('all.models').

prop.missing.cutoff

The maximum proportion of missing values allowed for each required gene.

validate.data

TRUE/FALSE, whether to validate input data. This should generally always be TRUE, but developers may set to FALSE to speedup testing/development.

Value

A list with the following objects:

  • features a data frame with predicted features as columns and patients as rows

  • validation results from validation checks telling you which features you were able to predict. See validate.gene.methy.data for more details.

Examples

data('all.models');

### example gene-level methylation data
data('example.data.gene.methy');
# note this dataset is derived from the following commands:
# data('example.data');
# example.data.gene.methy <- gene.methylation(example.data);

features <- estimate.features(example.data.gene.methy, all.models);
str(features$features);
#> 'data.frame':	2 obs. of  14 variables:
#>  $ age.continuous      : num  61.5 60.7
#>  $ ISUP.grade          : Factor w/ 5 levels "ISUP1","ISUP2",..: 3 2
#>  $ t.category          : Factor w/ 4 levels "T1","T2","T3",..: 3 2
#>  $ psa.categorical     : Factor w/ 3 levels "psa_lt_10","psa_10_19.9",..: 1 1
#>  $ pga                 : num  25.6 10.9
#>  $ CHD1.cna.loss       : Factor w/ 2 levels "Yes","No": 2 2
#>  $ NKX3.1.cna.loss     : Factor w/ 2 levels "Yes","No": 2 2
#>  $ MYC.cna.gain        : Factor w/ 2 levels "Yes","No": 1 2
#>  $ PTEN.cna.loss       : Factor w/ 2 levels "Yes","No": 2 2
#>  $ CDKN1B.cna.loss     : Factor w/ 2 levels "Yes","No": 2 2
#>  $ RB1.cna.loss        : Factor w/ 2 levels "Yes","No": 1 2
#>  $ CDH1.cna.loss       : Factor w/ 2 levels "Yes","No": 2 2
#>  $ TP53.cna.loss       : Factor w/ 2 levels "Yes","No": 2 2
#>  $ log2p1.snvs.per.mbps: num  1.026 0.545