I was trying to generate a figure based on the concentration vs absorbance. I have the absorbance from 4 fractions. All the values are numeric. I was trying to convert it into factor but it appeared as error like this -
data = read_excel("df.xlsx", StringsAsFactors = TRUE)
The output was
Error in read_excel("df.xlsx", StringsAsFactors = TRUE) : unused argument (StringsAsFactors = TRUE)
My questions are-
1. Do I need to convert the numeric value as factor for plotting?
2. Which argument should I use to plot multiple Fractions against the concentration?
1. readxl::read_excel() doesn't have the argument StringsAsFactors (see https://readxl.tidyverse.org/reference/read_excel.html).
If you want to convert your string data into factor, you can always do so later on. For example:
data %
dplyr::mutate(fractions = factor(fractions))
2. If you want to plot multiple fractions, you need to decide how your plot will look like. Will all 4 be plotted in the same graph, or you want to separate them into 4 subplots?
I guess both concentration and absorbance are continuous, then a scatter plot is suitable. You can use aes(color = fractions) to color the different fractions, or use facet_wrap(vars(fractions)) to make 4 separate subplots.
See guide here:
- https://r-charts.com/correlation/scatter-plot-group-ggplot2/
- https://ggplot2.tidyverse.org/reference/facet_wrap.html
Dear Long Bao Hoang , thanks a lot for your response. I want to plot everything in a single graph. I am attaching the file, can you please have a look and have suggestions on how can I use it to plot Concentration vs Absorbance for the line graph? (the data is not the actual one)
Dear Long Bao Hoang , I have figured out how should I plot that I asked you in the previous comment. However, do you think the plot is okay or I need to work on it (attached)? Because when the concentration is 0, shouldn't the line for the absorbance will be attached to the y axis?
Your plot looks great. If you want the X-axis to begin at 0, you can use scale_x_continuous(limits = c(0, NA), expand = c(0, 0)).
See my example code attached.
Dear Long Bao Hoang , thanks again. One more thing, since it's a bioassay result, I need to take the average value of three different values for a single fraction. In that case, do I need to put the SD in the graph? If yes, then how? I am attaching another file for your better understanding.
I believe this is more of a convention of your field, rather than a "must" or "must not". In epidemiology (my field), we tend to use standard errors (SE) instead of SD. If you have three samples, then SE = SD / sqrt(3).
You can easily calculate the SE using your data, then use geom_errorbar() to plot an error bar for each. See this: http://www.cookbook-r.com/Graphs/Plotting_means_and_error_bars_(ggplot2)/
You already have seen that there were four fractions, Fraction 1, Fraction 2, Fraction 3, and Fraction 4. Each of these fractions is from the average of three different values. It's like the average of "A, B, and C" is Fraction 1, "D, E, and F" is Fraction 2...like this. In this case, I will get four different SE from four different Fractions, isn't it?
That's correct. Basically you'll add the error bars to all the points in your above plot.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation