Function to predict using SuperSig object

Using a generated SuperSig, predict on a new dataset and return predicted probabilities for each observation.

predict_signature(object, newdata, factor)

Arguments

object	an object of class SuperSig
newdata	a data frame of mutations containing columns for sample_id, age, IndVar, and the 96 trinucleotide mutations (see vignette for details)
factor	the factor/exposure (e.g. "age", "smoking")

Value

predict_signature returns the original data frame with additional columns for the feature counts and classification score

Examples

head(example_dt) # use example data from package
#>   sample_id age chromosome  position ref alt
#> 1         1  50       chr1  94447621   G   C
#> 2         1  50       chr2 202005395   A   C
#> 3         1  50       chr7  20784978   T   A
#> 4         1  50       chr7  87179255   C   G
#> 5         1  50      chr19   1059712   G   T
#> 6         2  55       chr1  76226977   T   C
input_dt <- make_matrix(example_dt) # convert to correct format
input_dt$IndVar <- c(1, 1, 1, 0, 0) # add IndVar column
out <- get_signature(data = input_dt, factor = "Age") # get SuperSig
#> Begin feature engineering...
#> Begin cross-validated selection over 3 features and 15 inner folds...
#> ...testing inner fold 1
#> ...testing inner fold 2
#> ...testing inner fold 3
#> ...testing inner fold 4
#> ...testing inner fold 5
#> ...testing inner fold 6
#> ...testing inner fold 7
#> ...testing inner fold 8
#> ...testing inner fold 9
#> ...testing inner fold 10
#> ...testing inner fold 11
#> ...testing inner fold 12
#> ...testing inner fold 13
#> ...testing inner fold 14
#> ...testing inner fold 15
newdata <- predict_signature(out, newdata = input_dt, factor = "age")
suppressPackageStartupMessages({library(dplyr)})
head(newdata %>% select(score))
#> # A tibble: 5 x 1
#>   score
#>   <dbl>
#> 1   0.6
#> 2   0.6
#> 3   0.6
#> 4   0.6
#> 5   0.6