A function that computes the marker combinations and counts their corresponding positive samples for each class (once thresholds are selected).
Arguments
- data
a data.frame returned by load_data().
- signalthr
a numeric that specifies the value above which a marker expression is considered positive in a given sample. Since the target of the analysis is the identification of marker combinations capable to correctly classify samples, the user should choose a signalthr that:
Positively selects most samples belonging to the case class, which must be above signalthr.
Negatively selects most control samples, which must be below signalthr.
- combithr
a numeric that specifies the necessary number of positively expressed markers (>= signalthr), in a given combination, to cosinder that combination positivelly expressed in a sample.
- max_length
an integer that specifies the max combination length that is allowed
- case_class
a character that specifies which of the two classes of the dataset is the case class
Value
a data.frame containing how many samples of each class are "positive" for each combination, sensitivity and specificity.
Details
This function counts how many samples are 'positive' for each combination. A sample, to be considered positive for a given combination, must have a value higher than a given signal threshold (signalthr) for at least a given number of markers composing that combination (combithr).
Examples
if (FALSE) {
demo_data # combiroc built-in demo data (proteomics data from Zingaretti et al. 2012 - PMC3518104)
# To compute the marker combinations and count their corresponding positive samples for each class.
combs <- combi(data= demo_data, signalthr=450, combithr=1, case_class='A')
# count as positive the samples with value >= 450 for at least 1 marker in the combination
}