IPSSMmain: Molecular International Prognostic Scoring System IPSS-M

In papaemmelab/ipssm: Molecular International Prognosis Scoring System For Myelodysplastic Syndromes

Molecular International Prognostic Scoring System IPSS-M

Description

Main function of the IPSS-M model. It applies the IPSS-M on the patient processed clinical and molecular variables. It calculates the IPSS-M risk score and risk categories under the best, mean, and worst scenarios if some input data are missing. If no missing data, all scenarios are equal.

Usage

IPSSMmain(
  patientProcess,
  betaValues = c(HB1 = -0.171, TRANSF_PLT100 = -0.222, BLAST5 = 0.352, CYTOVEC = 0.287,
    TP53multi = 1.18, FLT3 = 0.798, MLL_PTD = 0.798, SF3B1_5q = 0.504, NPM1 = 0.43, RUNX1
    = 0.423, NRAS = 0.417, ETV6 = 0.391, IDH2 = 0.379, CBL = 0.295, EZH2 = 0.27, U2AF1 =
    0.247, SRSF2 = 0.239, DNMT3A = 0.221, ASXL1 = 0.213, KRAS = 0.202, SF3B1_alpha =
    -0.0794, nRes2 = 0.231),
  meanValues = c(HB1 = 9.87, TRANSF_PLT100 = 1.41, BLAST5 = 0.922, CYTOVEC = 1.39,
    TP53multi = 0.071, FLT3 = 0.0108, MLL_PTD = 0.0247, SF3B1_5q = 0.0166, NPM1 = 0.0112,
    RUNX1 = 0.126, NRAS = 0.0362, ETV6 = 0.0216, IDH2 = 0.0429, CBL = 0.0473, EZH2 =
    0.0588, U2AF1 = 0.0866, SRSF2 = 0.158, DNMT3A = 0.161, ASXL1 = 0.252, KRAS = 0.0271,
    SF3B1_alpha = 0.186, nRes2 = 0.388),
  bestValues = c(HB1 = 20, TRANSF_PLT100 = 2.5, BLAST5 = 0, CYTOVEC = 0, TP53multi = 0,
    FLT3 = 0, MLL_PTD = 0, SF3B1_5q = 0, NPM1 = 0, RUNX1 = 0, NRAS = 0, ETV6 = 0, IDH2 =
    0, CBL = 0, EZH2 = 0, U2AF1 = 0, SRSF2 = 0, DNMT3A = 0, ASXL1 = 0, KRAS = 0,
    SF3B1_alpha = 1),
  worstValues = c(HB1 = 4, TRANSF_PLT100 = 0, BLAST5 = 4, CYTOVEC = 4, TP53multi = 1,
    FLT3 = 1, MLL_PTD = 1, SF3B1_5q = 1, NPM1 = 1, RUNX1 = 1, NRAS = 1, ETV6 = 1, IDH2 =
    1, CBL = 1, EZH2 = 1, U2AF1 = 1, SRSF2 = 1, DNMT3A = 1, ASXL1 = 1, KRAS = 1,
    SF3B1_alpha = 0),
  rounding = TRUE,
  rounding.digits = 2,
  risk.cutpoints = c(-1.5, -0.5, 0, 0.5, 1.5),
  risk.cat = c("Very Low", "Low", "Moderate Low", "Moderate High", "High", "Very High")
)

Arguments

patientProcess	a patient processed data.frame, as the result of IPSSMprocess function.
betaValues	a covariate vector of the model weights. Should have name attributes.
meanValues	vector of average values for each covariate. Should have the same names attributes as in betaValues.
bestValues	vector of best values (leading to minimal risk) for each covariate (nRes2 not needed as already taken care of in IPSSMprocess).
worstValues	vector of worst values (leading to maximal risk) for each covariate (nRes2 not needed as already taken care of in IPSSMprocess).
rounding	should the raw IPSS-M risk score be rounded. Default is TRUE.
rounding.digits	number of digits for the rounding. Default is 2.
risk.cutpoints	cutpoints to be applied to the IPSS-M risk score to create risk categories.
risk.cat	names of the IPSS-M risk categories

Value

A result patient data.frame, same number of rows/patients as in patientProcess, with additonal columns labelled IPSSMscore_best, IPSSMscore_mean, IPSSMscore_worst, IPSSMcat_best, IPSSMcat_mean, IPSSMcat_worst, corresponding to the IPSS-M risk score and category in the best, mean, and worst scenarios.

Examples

dd <- read.csv(system.file("extdata", "IPSSMexample.csv", package = "ipssm"), header = TRUE)
dd.process <- IPSSMprocess(patientInput = dd)
dd.res <- IPSSMmain(patientProcess = dd.process)
print(dd.res[, c(1, grep("IPSSM", colnames(dd.res)))])

papaemmelab/ipssm documentation built on Feb. 8, 2023, 3:09 p.m.