Predict the exposure status of an independent sample.
pred(comb, indpt, names, field, count, copyrange)
| comb | List containing both negtive (n) and positive (v) clonotype percentages. |
|---|---|
| indpt | Vector of independent samples file paths. |
| names | Vector of labels for independent samples. |
| field | String containing the column or columns (space-delimited) of interest. |
| count | String containing the column name for colontype counts. |
| copyrange | Integer Vector of the min and max copy of a sequence, within a sample, to be considered. |
Matrix with % correct predictions from training data.
FIELD <- "vGeneName aminoAcid jGeneName" COUNT <- "copy" P_CUTOFF <- 0.1 MIN_PUBLIC <- 2 COPY_RANGE <- "1 99" listPos <- tsvDir(system.file("extdata", "Post", package="iCAT")) listNeg <- tsvDir(system.file("extdata", "Pre", package="iCAT")) naive <- readTrn(listNeg, FIELD, COUNT, COPY_RANGE, "naive") vaccs <- readTrn(listPos, FIELD, COUNT, COPY_RANGE, "vacc") mod <- train(naive, vaccs, listNeg, listPos, FIELD, COUNT, COPY_RANGE, P_CUTOFF, MIN_PUBLIC, NULL) pred(mod, system.file("extdata", "Pre", "KJW100_HLA-A2_10_PRE.tsv", package="iCAT"), "unknown-sample-label", FIELD, COUNT, COPY_RANGE)#> Sample Prediction % TARS #> [1,] "unknown-sample-label" "Negative" "0"