Write a summary() to as.data.frame for use in ggplot / R

Question

Please find af data sample t below.

I am conducting a competing risk analysis using etmCIF from the etm package - yielding the following, which is nice but needs better graphics:

There used to be a ggtrans.etm function to import data to ggplot. However, this function is apparently removed?!

However, I wish to transform my summary() into a data.frame but I receive an error:

library(etm)
cum_in <- etmCIF(Surv(os, event %in% c(1,2)) ~ 1, t, etype = event, failcode = c(1,2))
summary(cum_in)

Which gives

CIF 1 
          P        time         var      lower     upper n.risk n.event
 0.00000000   0.3297396 0.000000000 0.00000000 0.0000000    100       0
 0.00000000  57.5268750 0.000000000 0.00000000 0.0000000     90       0
 0.00000000 178.0340104 0.000000000 0.00000000 0.0000000     54       0
 0.06387317 271.0966667 0.001897498 0.01643949 0.2311213     22       0
 0.21669472 369.4858854 0.007605761 0.09511485 0.4494356     11       1
 0.21669472 925.1224479 0.007605761 0.09511485 0.4494356      2       0

CIF 2 
          P        time          var       lower     upper n.risk n.event
 0.01000000   0.3297396 0.0000990000 0.001414712 0.0688628    100       1
 0.07065711  57.5268750 0.0006633366 0.034315233 0.1425376     90       1
 0.14846026 178.0340104 0.0015118082 0.087973840 0.2445705     54       1
 0.23751402 271.0966667 0.0031735841 0.146981679 0.3703251     22       1
 0.23751402 369.4858854 0.0031735841 0.146981679 0.3703251     11       0
 0.56839997 925.1224479 0.0281468521 0.287757542 0.8751468      2       1

I need P, time, lower and upper in a data frame for ggplot2, so I tried

library(ggplot2)
ggplot(as.data.frame(cum_in), aes(x=time, y=P))  +
  geom_ribbon(data=cum_in, aes(ymin=lower, ymax=upper))

Which gives

Error in as.data.frame.default(cum_in) : cannot coerce class ‘"etmCIF"’ to a data.frame

Any idea how to transform summary() into something useful for ggplot? I would prefer not to downgrade the package.

UPDATED QUESTION

So I tried the function by @PoGibas, which worked nice initially. However, there seems to be a problem with the function.

I have updated the data sample t below

I have three cum. inc.-curves stratified for t$ki67in which constitute three different groups.

The cum. inc.-curves are estimated as follow

library(etm)
cum_in <- etmCIF(Surv(event.tid, event!=0) ~ ki67in, t, etype = event, failcode = 2)

In which plot(cum_in) correctly plots the following:

But when I try (based on the function etm_to_df)

res <- etm_to_df(cum_in)
ggplot(res, aes(time, P)) + 
  geom_ribbon(aes(ymin = lower, ymax = upper, fill = CIF), alpha = 0.2) +
  geom_line(aes(color = CIF))

I get this nonsense in ggplot (which does not seem to have three groups):

> head(res)
   CIF           P     time          var       lower      upper n.risk n.event
1: 0 1 0.009259259  0.25000 8.494005e-05 0.001309500 0.06390547    108       1
2: 0 1 0.018605870  1.75000 1.698800e-04 0.004685795 0.07234945    106       1
3: 0 1 0.028419811 11.83333 2.618497e-04 0.009249879 0.08556618    100       1
4: 0 1 0.028419811 12.00000 2.618497e-04 0.009249879 0.08556618     99       0
5: 0 1 0.028419811 15.00000 2.618497e-04 0.009249879 0.08556618     97       0
6: 0 1 0.038334927 18.00000 3.538387e-04 0.014552186 0.09898410     96       1
> tail(res)
   CIF          P     time         var      lower     upper n.risk n.event
1: 0 1 0.12156863 56.00000 0.006511402 0.03179904 0.4054164      9       0
2: 0 1 0.38184459 96.66667 0.049327707 0.10529823 0.8750079      3       1
3: 0 2 0.00000000  1.50000 0.000000000 0.00000000 0.0000000     17       0
4: 0 2 0.00000000  3.00000 0.000000000 0.00000000 0.0000000     15       0
5: 0 2 0.09760349 56.00000 0.008548335 0.01442923 0.5160136      9       1
6: 0 2 0.09760349 96.66667 0.008548335 0.01442923 0.5160136      3       0

My data sample

    t <- structure(list(ki67in = structure(c(0, 2, 0, 0, 1, 0, 2, 2, 1, 
0, 1, 2, 0, 2, 0, 1, 1, 1, 0, 2, 2, 0, 2, 1, 0, 0, 0, 1, 0, 1, 
2, 0, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 
0, 0, 0, 1, 0, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 1, 0, 1, 
0, 0, 1, 0, 0, 1, 2, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 2, 1, 2, 0, 2, 0, 0, 
1, 0, 0, 0, 0, 0, 0, 1, 0, 2, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 
0, 2, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 
0, 0, 0), class = "AsIs"), event = structure(c(1, 1, 1, 1, 1, 
0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 2, 0, 2, 
0, 0, 2, 0, 0, 1, 0, 2, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 2, 2, 0, 
0, 0, 2, 0, 0, 0, 2, 2, 0, 2, 1, 0, 2, 0, 2, 0, 2, 0, 0, 0, 1, 
0, 1, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 2, 2, 0, 0, 0, 2, 0, 1, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 
0, 0, 2, 0, 0, 0, 0), class = "AsIs"), event.tid = c(1.75, 1.5, 
11.83333333, 0.25, 1.75, 1, 2, 96.66666667, 2, 106.5833333, 3, 
3, 3, 4, 4, 4, 141.9166667, 5, 6, 7, 8, 8, 8, 9, 11, 12, 13, 
13, 15, 15, 15, 40.91666667, 17, 17, 18, 173, 28, 29, 30, 33, 
34, 35, 178.5833333, 37, 38, 39, 40, 41, 45, 49, 49, 50, 52, 
53, 54, 56, 56, 194.4166667, 56, 57, 58, 58, 60, 60, 60, 60, 
61, 275.75, 63, 189.75, 66, 67, 67, 72, 72, 74, 78, 80, 80, 80, 
81, 82, 83, 83, 84, 84, 85, 85, 86, 86, 88, 88, 88, 88, 89, 89, 
89, 90, 90, 91, 91, 92, 92, 251.8333333, 92, 93, 93, 93, 93, 
93, 93, 94, 97, 98, 98, 99, 99, 99, 100, 101, 101, 101, 103, 
103, 103, 103, 104, 104, 106, 106, 109, 110, 111, 111, 112, 114, 
114, 115, 116, 117, 299.8333333, 118, 118, 119, 120, 120, 120, 
120, 120, 120, 121, 121, 123, 124, 124, 125, 125, 125, 125)), class = "data.frame", row.names = c(1L, 
2L, 3L, 9L, 10L, 11L, 12L, 13L, 14L, 15L, 16L, 18L, 19L, 20L, 
21L, 22L, 23L, 24L, 25L, 26L, 27L, 28L, 29L, 30L, 31L, 32L, 33L, 
34L, 35L, 36L, 37L, 38L, 39L, 40L, 41L, 44L, 45L, 46L, 47L, 48L, 
49L, 50L, 51L, 52L, 53L, 54L, 55L, 57L, 59L, 60L, 61L, 62L, 63L, 
64L, 65L, 66L, 67L, 68L, 69L, 70L, 71L, 72L, 73L, 74L, 75L, 76L, 
77L, 78L, 79L, 80L, 81L, 82L, 83L, 84L, 85L, 87L, 89L, 90L, 91L, 
92L, 93L, 94L, 96L, 97L, 98L, 99L, 100L, 101L, 102L, 103L, 104L, 
105L, 106L, 107L, 109L, 110L, 111L, 112L, 113L, 114L, 115L, 116L, 
117L, 118L, 119L, 120L, 121L, 123L, 124L, 125L, 126L, 127L, 128L, 
130L, 131L, 132L, 133L, 134L, 135L, 136L, 137L, 138L, 139L, 140L, 
141L, 142L, 143L, 144L, 145L, 146L, 147L, 148L, 149L, 150L, 151L, 
152L, 153L, 154L, 155L, 156L, 157L, 158L, 159L, 160L, 161L, 162L, 
163L, 164L, 165L, 166L, 167L, 168L, 169L, 170L, 171L, 172L, 173L, 
174L, 175L))

user_1286528 · Answer

ggtransfo.etm was removed with this commit. One way is to play around that function, but I tried to improve etm:::summary.etmCIF function to return binded data frame (introduced data.table as a dependency):

# NEW VERSION (adapted according to question update)
# Works with multiple groups 
etm_to_df <- function(object, ci.fun = "cloglog", level = 0.95, ...) {
  l.X <- ncol(object$X)
  l.trans <- nrow(object[[1]]$trans)
  res <- list()
  for (i in seq_len(l.X)) {
      temp <- summary(object[[i]], ci.fun = ci.fun, level = level)
      res[[i]] <- data.table::rbindlist(
        temp[object$failcode + 1], idcol = "CIF"
      )[, CIF := paste0("CIF", CIF, "; ", names(object)[i])]
  }
  do.call(rbind, res)
}

This function returns a data frame with column CIF which contains identifier.

# With given OPs data one can use 
library(etm)
cum_in <- etmCIF(Surv(os, event %in% c(1,2)) ~ 1, n, etype = event, failcode = c(1,2))
res <- etm_to_df(cum_in)

Then it's easy to plot it using ggplot2:

library(ggplot2)
ggplot(res, aes(time, P)) +
  geom_ribbon(aes(ymin = lower, ymax = upper, fill = CIF), alpha = 0.2) +
  geom_line(aes(color = CIF)) +
  scale_fill_manual(values = c("red", "blue")) +
  scale_color_manual(values = c("red", "blue")) +
  theme_classic()

Old function:

# Same functionality as etm:::summary.etmCIF, but returns a data frame
etm_to_df <- function(object, ci.fun = "cloglog", level = 0.95, ...) {
  l.X <- ncol(object$X)
  l.trans <- nrow(object[[1]]$trans)
  temp <- lapply(object[seq_len(l.X)], function(ll) {
    res <- summary(ll, ci.fun = ci.fun, level = level, ...)
    data.table::rbindlist(res[seq_len(l.trans) + 1], idcol = "CIF")
  })
  do.call(rbind, temp)
}

user_9993361 · Answer

You can assign the summary to a variable and then get direct access like this:

s <- summary(cum_in)
s[[1]]$`CIF 1`
s[[1]]$`CIF 2`

A basic ggplot could look like this

library(ggplot2)
ggplot(s[[1]]$`CIF 1`, aes(x = time, y = P))  +
  geom_step() + 
  geom_step(data = s[[1]]$`CIF 2`, aes(x = time, y = P), lty = 2)

Write a summary() to as.data.frame for use in ggplot / R

2 Answers