DLBCL                package:maxstat                R Documentation

_D_i_f_f_u_s_e _l_a_r_g_e _B-_c_e_l_l _l_y_m_p_h_o_m_a

_D_e_s_c_r_i_p_t_i_o_n:

     A data frame with gene expression data from DLBCL (diffuse large
     B-cell lymphoma) patients.

_U_s_a_g_e:

     data(DLBCL)

_F_o_r_m_a_t:

     `_D_L_C_L_i_d' DLCL identifier

     `_G_E_G' Gene Expression Group

     `_t_i_m_e' survival time in month

     `_c_e_n_s' censoring: 0 cencored, 1 dead

     `_I_P_I' International Prognostic Index

     `_M_G_E' Mean Gene Expression

     Except of `MGE', the data is published at <URL:
     http://llmpp.nih.gov/lymphoma/data.shtml>. `MGE' was computed by
     Berthold Lausen.

_S_o_u_r_c_e:

     Ash A. Alizadeh et. al (2000), Distinct types of diffuse large
     B-cell lymphoma identified by gene expression profiling. Nature,
     403, 504-509

_R_e_f_e_r_e_n_c_e_s:

     Lausen, B. and Schumacher, M. (1992), Maximally Selected Rank
     Statistics. Biometrics 48, 73-85

_E_x_a_m_p_l_e_s:

     data(DLBCL)

     # remove NA's

     DLBCL <- DLBCL[!is.na(DLBCL$time),]

     # compute the cutpoint

     postscript("statDLBCL.ps",horizontal=F, width=8, height=8)
     par(mai=c(1.0196235, 1.0196235, 0.8196973, 0.4198450))

     mod <- maxstat.test(DLBCL$MGE, DLBCL$time,
                  cens=DLBCL$cens, smethod="LogRank", plot=T, cex.lab=1.6,
                  cex.axis=1.6, xlab="Mean gene expression")

     dev.off()

     # significance of the cutpoint
     # Limiting distribution

     maxstat.test(DLBCL$MGE, DLBCL$time,
                  cens=DLBCL$cens, smethod="LogRank", pmethod="Lau92")

     # improved Bonferroni inequality

     maxstat.test(DLBCL$MGE, DLBCL$time,
                  cens=DLBCL$cens, smethod="LogRank", pmethod="Lau94")

     # small sample solution Hothorn & Lausen (2001)

     maxstat.test(DLBCL$MGE, DLBCL$time,
                  cens=DLBCL$cens, smethod="LogRank", pmethod="HL")

     maxstat.test(DLBCL$MGE, DLBCL$time,
                  cens=DLBCL$cens, smethod="LogRank", pmethod="exactGauss")

     # Nature article survival analysis

     splitGEG <- rep(1, nrow(DLBCL))
     DLBCL <- cbind(DLBCL, splitGEG)
     DLBCL$splitGEG[DLBCL$GEG == "Activated B-like"] <- 0

     plot(survfit(Surv(time, cens) ~ splitGEG, data=DLBCL),
          xlab="Survival time in month", ylab="Probability")

     text(90, 0.7, "GC B-like")
     text(60, 0.3, "Activated B-like")

     splitIPI <- rep(1, nrow(DLBCL))
     DLBCL <- cbind(DLBCL, splitIPI)
     DLBCL$splitIPI[DLBCL$IPI <= 2] <- 0

     plot(survfit(Surv(time, cens) ~ splitIPI, data=DLBCL),
          xlab="Survival time in month", ylab="Probability")

     text(90, 0.7, "Low clinical risk")
     text(60, 0.25, "High clinical risk")

     # survival analysis using the cutpoint 

     splitMGE <- rep(1, nrow(DLBCL))
     DLBCL <- cbind(DLBCL, splitMGE)
     DLBCL$splitMGE[DLBCL$MGE <= mod$estimate] <- 0

     postscript("survDLBCL.ps",horizontal=F, width=8, height=8)
     par(mai=c(1.0196235, 1.0196235, 0.8196973, 0.4198450))

     plot(survfit(Surv(time, cens) ~ splitMGE, data=DLBCL),
     xlab = "Survival time in month",
     ylab="Probability", cex.lab=1.6, cex.axis=1.6)

     text(90, 0.9, expression("Mean gene expression" > 0.186), cex=1.6)   
     text(90, 0.45, expression("Mean gene expression" <= 0.186 ), cex=1.6)   

     dev.off()

