rfacet-wrapmanipulate

How to set up manipulate package

I am trying to figure out to analyze sepated group of different boxplot from the same matrix and avoid facet_wrap2 (too many data in there) I would like visualize each amino acid independently e.i C8 only, then C8_IS, etc.. or even by group of two amino such as C8+C8_IS and then scroll into Orn + ORN_IS

folling script:

ggboxplot(intensity_test3, x = "sample", y = "Intensity", color = "day", palette = "lancet",
add="jitter", shape = "sample") + facet_grid2(~analyte~conc+method, margins = "vs", scales = "free")+
 theme_bw(base_size = 18)  +
 theme(strip.text.y = element_text(angle = 90))  + scale_y_continuous(labels = function(x) format(x, scientific = TRUE, accuracy = 1))

enter image description here

Solution

  • So I just made up some data to match your graph and code. Here it is:

    Just run this code to get the dataset I used.

    intensity_test3 <- structure(list(sample = c("Low_qc", "Low_qc", "Low_qc", "Low_qc", 
"Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", 
"Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", 
"Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", 
"Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", "Low_qc", 
"high_qc", "high_qc", "high_qc", "high_qc", "high_qc", "high_qc", 
"high_qc", "high_qc", "high_qc", "high_qc", "high_qc", "high_qc", 
"high_qc", "high_qc", "high_qc", "high_qc", "high_qc", "high_qc", 
"high_qc", "high_qc", "high_qc", "high_qc", "high_qc", "high_qc", 
"high_qc", "high_qc", "high_qc", "high_qc", "high_qc", "high_qc", 
"high_qc", "high_qc"), Intensity = c(1e+05, 1e+05, 1e+05, 1e+05, 
1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 
1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 
1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 1e+05, 
1e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 
2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 
2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 
2e+05, 2e+05, 2e+05, 2e+05, 2e+05, 2e+05), day = c("day_1", "day_1", 
"day_1", "day_1", "day_2", "day_2", "day_2", "day_2", "day_3", 
"day_3", "day_3", "day_3", "day_4", "day_4", "day_4", "day_4", 
"day_1", "day_1", "day_1", "day_1", "day_2", "day_2", "day_2", 
"day_2", "day_3", "day_3", "day_3", "day_3", "day_4", "day_4", 
"day_4", "day_4", "day_1", "day_1", "day_1", "day_1", "day_2", 
"day_2", "day_2", "day_2", "day_3", "day_3", "day_3", "day_3", 
"day_4", "day_4", "day_4", "day_4", "day_1", "day_1", "day_1", 
"day_1", "day_2", "day_2", "day_2", "day_2", "day_3", "day_3", 
"day_3", "day_3", "day_4", "day_4", "day_4", "day_4"), analyte = c("C8", 
"C8_I6", "orn", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "om", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "orn", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "om", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "orn", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "om", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "orn", "orn_IS", "C8", "C8_I6", "orn", "orn_IS", "C8", 
"C8_I6", "om", "orn_IS", "C8", "C8_I6", "orn", "orn_IS"), conc = c(2.7, 
2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 
2.7, 2.7, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 
8.1, 8.1, 8.1, 8.1, 8.1, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 
2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 2.7, 8.1, 8.1, 8.1, 8.1, 8.1, 
8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1, 8.1), method = c("new", 
"new", "new", "new", "new", "new", "new", "new", "new", "new", 
"new", "new", "new", "new", "new", "new", "old", "old", "old", 
"old", "old", "old", "old", "old", "old", "old", "old", "old", 
"old", "old", "old", "old", "new", "new", "new", "new", "new", 
"new", "new", "new", "new", "new", "new", "new", "new", "new", 
"new", "new", "old", "old", "old", "old", "old", "old", "old", 
"old", "old", "old", "old", "old", "old", "old", "old", "old"
)), class = "data.frame", row.names = c(NA, -64L))

    Then I made a for loop and filter to only run each analyte. It should print each figure separately. You'' have to use the arrows in your plot viewer to see them as they are generated (assuming you're using R studio). Let me know if you have any questions, or if that's what you were looking for.

    for (i in unique(intensity_test3$analyte)){
  p <- ggboxplot(intensity_test3 %>% dplyr::filter(analyte %in% c(i)), x = "sample", y = "Intensity", color = "day", palette = "lancet",
            add="jitter", shape = "sample") + facet_grid2(~analyte~conc+method, margins = "vs", scales = "free")+
    theme_bw(base_size = 18)  +
    theme(strip.text.y = element_text(angle = 90))  + scale_y_continuous(labels = function(x) format(x, scientific = TRUE, accuracy = 1))
  print(p)
}

    Here is an example of one figure it made.

    enter image description here

    EDIT:

    You can certainly choose which panels go together. You could just do it like this (simply filter the data you give ggboxplot):

    p1 <- ggboxplot(intensity_test3 %>% dplyr::filter(analyte %in% c("C8", "C8_I6")), x = "sample", y = "Intensity", color = "day", palette = "lancet",
                add="jitter", shape = "sample") + facet_grid2(~analyte~conc+method, margins = "vs", scales = "free")+
  theme_bw(base_size = 18)  +
  theme(strip.text.y = element_text(angle = 90))  + scale_y_continuous(labels = function(x) format(x, scientific = TRUE, accuracy = 1))

p2 <- ggboxplot(intensity_test3 %>% dplyr::filter(analyte %in% c("orn", "orn_IS")), x = "sample", y = "Intensity", color = "day", palette = "lancet",
                add="jitter", shape = "sample") + facet_grid2(~analyte~conc+method, margins = "vs", scales = "free")+
  theme_bw(base_size = 18)  +
  theme(strip.text.y = element_text(angle = 90))  + scale_y_continuous(labels = function(x) format(x, scientific = TRUE, accuracy = 1))

print(p1)
print(p2)

    As for the t-test, this might be better for a separate question on SO or even perhaps the "stats" version of SO, Cross Validated if needed. It sounds like you have multiple factors and I don't know exactly what questions you would want to answer. Making a new post, and taking the time to fully describe what statistics you would like to perform would be most helpful for you.