# Generate a `ggplot` for a `survfit` or `ten` object ```r autoplot(object, ...) ## S3 method for class 'ten' autoplot( object, ..., title = "Marks show times with censoring", type = c("single", "CI", "fill"), alpha = 0.05, ciLine = 10, censShape = 3, palette = c("Dark2", "Set2", "Accent", "Paired", "Pastel1", "Pastel2", "Set1", "Set3"), jitter = c("none", "noEvents", "all"), tabTitle = "Number at risk by time", xLab = "Time", timeTicks = c("major", "minor", "days", "months", "custom"), times = NULL, yLab = "Survival", yScale = c("perc", "frac"), legend = TRUE, legTitle = "Group", legLabs = NULL, legOrd = NULL, titleSize = 15, axisTitleSize = 15, axisLabSize = 10, survLineSize = 0.5, censSize = 5, legTitleSize = 10, legLabSize = 10, fillLineSize = 0.05, tabTitleSize = 15, tabLabSize = 5, nRiskSize = 5 ) ## S3 method for class 'stratTen' autoplot( object, ..., title = NULL, type = c("single", "CI", "fill"), alpha = 0.05, ciLine = 10, censShape = 3, palette = c("Dark2", "Set2", "Accent", "Paired", "Pastel1", "Pastel2", "Set1", "Set3"), jitter = c("none", "noEvents", "all"), tabTitle = "Number at risk by time", xLab = "Time", timeTicks = c("major", "minor", "days", "months", "custom"), times = NULL, yLab = "Survival", yScale = c("perc", "frac"), legend = TRUE, legTitle = "Group", legLabs = NULL, legOrd = NULL, titleSize = 15, axisTitleSize = 15, axisLabSize = 10, survLineSize = 0.5, censSize = 5, legTitleSize = 10, legLabSize = 10, fillLineSize = 0.05, tabTitleSize = 15, tabLabSize = 5, nRiskSize = 5 ) ## S3 method for class 'survfit' autoplot( object, ..., title = "Marks show times with censoring", type = c("single", "CI", "fill"), alpha = 0.05, ciLine = 10, censShape = 3, palette = c("Dark2", "Set2", "Accent", "Paired", "Pastel1", "Pastel2", "Set1", "Set3"), jitter = c("none", "noEvents", "all"), tabTitle = "Number at risk by time", xLab = "Time", timeTicks = c("major", "minor", "weeks", "months", "custom"), times = NULL, yLab = "Survival", yScale = c("perc", "frac"), legend = TRUE, legLabs = NULL, legOrd = NULL, legTitle = "Group", titleSize = 15, axisTitleSize = 15, axisLabSize = 10, survLineSize = 0.5, censSize = 5, legTitleSize = 10, legLabSize = 10, fillLineSize = 0.05, tabTitleSize = 15, tabLabSize = 5, nRiskSize = 5, pVal = FALSE, sigP = 1, pX = 0.1, pY = 0.1 ) ``` ## Arguments - `object`: An object of class `survfit`, `ten` or `stratTen`. - `...`: Additional arguments (not implemented). - `title`: Title for survival plot. - `type`: `type="single"` (the default) plots single lines. - **`type="CI"`**: Adds lines indicating confidence intervals (taken from `upper` and `lower` values of `survfit` object). Higher values of `alpha` (opacity) are recommended for this, e.g. `alpha=0.8`. - **`type="fill"`**: Adds filled rectangles from the survival lines to the confidence intervals above. - `alpha`: Opacity of lines indicating confidence intervals or filled rectangles. Should be in range $0-1$. Lower = more transparent. Larger values e.g. `alpha=0.7` are recommended for confidence intervals. - `ciLine`: Confidence interval line type. See 'line type specification' in ?graphics::par - `censShape`: Shape of marks to indicate censored onservations. Default is `3` which gives vertical ticks. Use `censShape=10` for circular marks. See ?graphics::points - `palette`: Options are taken from [color_brewer](https://colorbrewer2.org/). * `palette="Dark2"` (the default) is recommended for `single` or `CI` plots. * `palette="Set2"` is recommended for `type="fill"` plots. - `jitter`: By default, `jitter="none"`. * If `jitter="noEvents"`, adds some random, positive noise to survival lines with no events (i.e. all observations censored). This will bring them just above 1 on the y-axis, making them easier to see separately. * If `jitter="all"` add some vertical and horizontal noise to all survival lines. This can prevent overlapping of lines for censoring. - `tabTitle`: Table title. -Axis arguments: - `xLab`: Label for $x$ axis on survival plot. - `timeTicks`: Numbers to mark on the $x$ axis of the survival plot and the table. - **`"major"`**: (the default) only the major $x$-axis (time) marks from the survival plot are are labelled on the plot and table. - **`"minor"`**: minor axis marks are labelled instead. - **`"days"`**: scale is $0, 7, 14, ..., t_{max}$ - **`"months"`**: scale is $0, 12,, 24, ..., t_{max}$ - **`"custom"`**: scale is given by `times` below - `times`: Vector of custom times to use for $x$ axis. - `yLab`: Label for $y$ axis on survival plot. - `yScale`: Display for point on $y$ axis: - **`"perc"`**: Displays as percentages. - **`"frac"`**: Displays as fractions e.g. $0, 0.1, 0.2, ..., 1.0.$ -Legend arguments: - `legend`: If `legend=FALSE`, no legends will be produced for the plot or table. - `legTitle`: Legend title. - `legLabs`: Legend labels. These can be used to replace the names of the covariate groups ('strata' in the case of a `survfit` object). Should be given in the same order as those strata. - `legOrd`: Legend order. -Size arguments: Size arguments are passed to `ggplot2::element_text(size=)`. - `titleSize`: Title size for survival plot. - `axisTitleSize`: Title size for axes. - `axisLabSize`: Title size for labels on axes. - `survLineSize`: Survival line size. - `censSize`: Size of marks to indicate censored onservations. - `legTitleSize`: Title size for legend. - `legLabSize`: Legend labels width and height. - `fillLineSize`: Line size surrouding filled boxes. - `tabTitleSize`: Table title text size. - `tabLabSize`: Table legend text size. - `nRiskSize`: Number at risk - text size. -Arguments for autoplot.survfit only: - `pVal`: If `pVal=TRUE`, adds $p$ value from log-rank test to plot - `sigP`: No. of significant digits to display in $p$ value. Typically $1$ to $3$. - `pX`: Location of $p$ value on $x$ axis. Should be in the range of $0 - 1$, where value is to be placed relative to the maximum observed time. E.g. `pX = 0.5` will place it half-way along $x$-axis - `pY`: Location of $p$ value on $y$ axis. Should be in the range of $0 - 1$, as above. ## Note `autoplot.survfit` may be deprecated after packageVersion 0.6. Please try to use `autoplot.ten` instead. ## Examples ```r ## examples are slow to run; see vignette for output from these ## Not run: ### autoplot.ten data("kidney", package="KMsurv") t1 <- ten(survfit(Surv(time, delta) ~ type, data=kidney)) autoplot(t1) autoplot(t1, type="fill", survLineSize=2, jitter="all") autoplot(t1, timeTicks="months", type="CI", jitter="all", legLabs=c("surgical", "percutaneous"), title="Time to infection following catheter placement \n by type of catheter, for dialysis patients", titleSize=10, censSize=2)$plot t2 <- ten(survfit(Surv(time=time, event=delta) ~ 1, data=kidney)) autoplot(t2, legLabs="")$plot autoplot(t2, legend=FALSE) data("rectum.dat", package="km.ci") t3 <- ten(survfit(Surv(time, status) ~ 1, data=rectum.dat)) ## change confidence intervals to log Equal-Precision confidence bands ci(t3, how="nair", tL=1, tU=40) autoplot(t3, type="fill", legend=FALSE)$plot ## manually changing the output t4 <- ten(survfit(Surv(time, delta) ~ type, data=kidney)) (a4 <- autoplot(t4, type="CI", alpha=0.8, survLineSize=2)$plot) ## change default colors a4 + list(ggplot2::scale_color_manual(values=c("red", "blue")), ggplot2::scale_fill_manual(values=c("red", "blue"))) ## change limits of y-axis suppressMessages(a4 + ggplot2::scale_y_continuous(limits=c(0, 1))) ## End(Not run) ## Not run: data("pbc", package="survival") t1 <- ten(Surv(time, status==2) ~ trt + strata(edema), data=pbc, abbNames=FALSE) autoplot(t1) ## End(Not run) ### autoplot.survfit ## Not run: data(kidney, package="KMsurv") s1 <- survfit(Surv(time, delta) ~ type, data=kidney) autoplot(s1, type="fill", survLineSize=2) autoplot(s1, type="CI", pVal=TRUE, pX=0.3, legLabs=c("surgical", "percutaneous"), title="Time to infection following catheter placement \n by type of catheter, for dialysis patients")$plot s1 <- survfit(Surv(time=time, event=delta) ~ 1, data=kidney) autoplot(s1, legLabs="")$plot autoplot(s1, legend=FALSE)$plot data(rectum.dat, package="km.ci") s1 <- survfit(Surv(time, status) ~ 1, data=rectum.dat) ## change confidence intervals to log Equal-Precision confidence bands if (require("km.ci")) { km.ci::km.ci(s1, method="logep") autoplot(s1, type="fill", legend=FALSE)$plot } ## manually changing the output s1 <- survfit(Surv(time, delta) ~ type, data=kidney) g1 <- autoplot(s1, type="CI", alpha=0.8, survLineSize=2)$plot ## change default colors g1 + ggplot2::scale_colour_manual(values=c("red", "blue")) + ggplot2::scale_fill_manual(values=c("red", "blue")) ## change limits of y-axis g1 + ggplot2::scale_y_continuous(limits=c(0, 1)) ## End(Not run) ``` ## See Also ?ggplot2::ggplot_build ## Author(s) Chris Dardis. `autoplot.survfit` based on existing work by R. Saccilotto, Abhijit Dasgupta, Gil Tomas and Mark Cowley.

autoplotTen function