[R] Fast way to draw mean values and 95% confidence intervals of groups with ggplot2

[Luigi Marongiu]

Thank you for your answer. I will implement it, but still I reckon
ggplot2 cannot do the whole thing on its own terms: I need to prep the
data beforehand.
Cheers
Luigi

On Thu, Nov 16, 2023 at 1:39 PM Rui Barradas <ruipbarradas using sapo.pt> wrote:
>
> Às 11:59 de 16/11/2023, Luigi Marongiu escreveu:
> > Hello,
> > I have triplicate (column A) readings (column D) of samples exposed to
> > different concentrations (column C) over time (column B).
> > Is it possible to draw a line plot of the mean values for each
> > concentration (C)? At the moment, I get a single line.
> > Also, is there a simple way to draw the 95% CI around these data? I
> > know I need to use ribbon with the lower and upper limit, but is there
> > a simple way for ggplot2 to calculate directly these values?
> > Here is a working example:
> >
> > ```
> > A = c(rep(1, 28), rep(2, 28), rep(3, 28))
> > B = rep(c(0, 15, 30, 45, 60, 75, 90), 12)
> > C = rep(c(rep(0, 7), rep(0.6, 7), rep(1.2, 7),
> >            rep(2.5,7)),3)
> > D = c(731.33,    761.67,    730,    761.67,    741.67,    788.67,    784.33,
> >        686.67,    685.33,    680,    693.67,    684,    704,    709.67,    739,
> >        731,    719,    767,    760.67,    776.67,    768.67,    675,    671.67,
> >        668.67,    677.33,    673.67,    687,    696.67,    727,    750.67,
> >        752.67,    786.67,    794.67,    843.33,    946,    732.67,    737.33,
> >        775.33,    828,    918,    1063,    1270,    752.67,    742.33,
> >    735.67,
> >        747.67,    777.33,    803.67,    865.67,    700,    700.67,    705.67,
> >        722.67,    744,    779,    837,    748,    742,    754,    747.67,
> >        775.67,    808.67,    869,    705.67,    714.33,    702.33,    730,
> >        710.67,    731,    744,    686.33,    687.33,    670,    702.33,
> >        669.33,    707.33,    708.33,    724,    747,    761.33,    715,
> >        697.67,    728,    728)
> >
> > df = data.frame(A, B, C, D)
> > library(ggplot2)
> > ggplot(data=df, aes(x=B, y=D, z=C, color =C)) +
> >    geom_line(stat = "summary", fun = "mean") +
> >    geom_ribbon()
> > ```
> >
> > Thank you
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> Hello,
>
> I am not sure that the code below is what you want.
> The first 3 instructions are to create a named vector of colors.
> The pipe is what tries to solve the problem. It computes means and se's
> by groups of time and concentration, then plots the ribbon below the lines.
>
> It is important to not set color = C in the initial call to ggplot,
> since it would be effective in all the subsequent layers (try it).
> To have one line per concentration I use group = C instead.
>
>
>
> suppressPackageStartupMessages({
>    library(ggplot2)
>    library(dplyr)
> })
>
> n_colors <- df$C |> unique() |> length()
> names_colors <- df$C |> unique() |> as.character()
> clrs <- setNames(palette.colors(n_colors), names_colors)
>
> df %>%
>    mutate(C = factor(C)) %>%
>    group_by(B, C) %>%
>    mutate(mean_D = mean(D), se_D = sd(D)) %>%
>    ungroup() %>%
>    ggplot(aes(x = B, group = C)) +
>    geom_ribbon(aes(ymin = mean_D - se_D, ymax = mean_D + se_D), fill =
> "grey", alpha = 0.5) +
>    geom_line(aes(y = mean_D, color = C)) +
>    geom_point(aes(y = D, color = C)) +
>    scale_color_manual(name = "Concentration", values = clrs)
>
>
> Hope this helps,
>
> Rui Barradas
>
>
> --
> Este e-mail foi analisado pelo software antivírus AVG para verificar a presença de vírus.
> www.avg.com



-- 
Best regards,
Luigi