Plots diagnostic and performance information for fits in a
MessinaSurvResult object, as returned by
the result of a Messina survival analysis, as
additional options to control the plot:
the y coordinates of points in the plot,
For each feature index given by indices, produces four plots:
A plot of the
value of the objective function over all possible
thresholds. Each sample is represented by a point on the
objective function trace. The selected threshold, if any,
is shown by a solid vertical line, and the margins by
dotted vertical lines on either side of it. The minimum
values of the objective function specified by the user are
shown as horizontal dotted lines. This plot is useful for
assessing fit stability, particularly for the "coxcoef" and
"reltau" objective functions, which can be unstable at low
or high threshold values. See
This Kaplan-Meier plot shows two traces, showing the outcomes of the two subgroups in the cohort defined by whether the plotted feature is above or below the threshold. Optionally (if bootstrap_type != "none"), the KM traces will be surrounded by shaded regions that represent either +/- 1 SD (bootstrap_type == "stdev") or a bootstrap_ci confidence interval (bootstrap_type == "ci").
This plot is identical to the above, except that the performance when the lower margin is used to separate the sample groups is shown.
This plot is identical to the above, except that the performance when the upper margin is used to separate the sample groups is shown. These last two plots give an indication of the robustness of the MessinaSurv fit at its extremes.
The Kaplan-Meier plots may optionally display bootstrap bands, if bootstrap_type != "none". Note that the calculation of bootstrap bands is computationally-intensive, and this function will by default use multiprocessing to speed calculations if doMC is loaded and more than one core registered for use. For examples of the plots and their interpretation, see the vignette.
Mark Pinese [email protected]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
## Load a subset of the TCGA renal clear cell carcinoma data ## as an example. data(tcga_kirc_example) ## Run the messinaSurv analysis on these data. Use a tau ## objective, with a minimum performance of 0.6. Note that ## messinaSurv analyses are very computationally-intensive, ## so multicore use with doMC loaded and parallel = TRUE is ## strongly recommended. In this example we use a single ## core by default. fit = messinaSurv(kirc.exprs, kirc.surv, obj_func = "tau", obj_min = 0.6) ## Plot the three best features found by Messina plot(fit, indices = 1:3) ## Plot the best feature found by Messina, with 90% confidence bands. ## Note that the bootstrap iterations can be slow, so it is ## recommended that multiple cores are used, with doMC loaded ## and parallel = TRUE. plot(fit, indices = 1, bootstrap_type = "ci", bootstrap_ci = 0.9) ## Plot the Messina fit of the 10th feature in the dataset, with ## +/- 1 standard deviation bands. plot(fit, indices = 10, sort_features = FALSE, bootstrap_type = "stdev")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.