Three Groups Example
on Monday, November 7th, 2022 12:06 | by Silvia Marcato
Pseudocode of the statical analysis with three groups according to the group descriptions.
# If n. of groups = 3 and n. of unique descriptions = 2, then perform
# the statistical analysis between singleton and each doubleton group.
# Else if n. of groups = 3 and descriptions are all identical/different,
# then perform the statistical analysis between each one of them.
if (NofGroups==3 & length(unique(groupdescriptions))==2) {
statistical_analysis(singleton, doubleton_1)
statistical_analysis(singleton, doubleton_2)
plot_results
} else {
statistical_analysis(group_1, group_2)
statistical_analysis(group_1, group_3)
statistical_analysis(group_2, group_3)
plot_results
}
Category: R code | No Comments
Three groups issue
on Monday, September 19th, 2022 1:00 | by Silvia Marcato
if(NofGroups == 3 & length(unique(groupdescriptions))==2){
doubleton <- list()
singleton <- list()
if (groupdescriptions[1] == groupdescriptions[2]) {
doubleton = c(unique(groupdescriptions[1], groupdescriptions[2]),
groupnames[1], groupnames[2])
singleton = c(groupdescriptions[3], groupnames[3])
} else if (groupdescriptions[2] == groupdescriptions[3]) {
doubleton = c(unique(groupdescriptions[2], groupdescriptions[3]),
groupnames[2], groupnames[3])
singleton = c(groupdescriptions[1], groupnames[1])
} else {
doubleton = c(unique(groupdescriptions[1], groupdescriptions[3]),
groupnames[1], groupnames[3])
singleton = c(groupdescriptions[2], groupnames[2])
}
}
Given three descriptions, splits them into three variables. Two out of three of these are the same while the other is not: the single set to be compared to the two “identical” sets is emplaced in the singleton list while the other two sets are emplaced in the doubleton list.
Category: R code | No Comments
Progress Week 29
on Monday, July 20th, 2020 1:38 | by Anders Eriksson
-Introduced Sayani to the wonderful world of Drosophila
-Been doing some DTS coding
-Preparing flies to do optomotor response for Mathias Raß

Category: flight, genetics, Lab, Memory, Optomotor response, R code | No Comments
Progress week 26-28
on Monday, July 13th, 2020 1:39 | by Anders Eriksson
Updates in DTS code

Refreshing dissection skills


Category: Anatomy, genetics, Lab, neuronal activation, personal, R code, science, Uncategorized | No Comments
Progress for week 25:
on Monday, June 22nd, 2020 1:58 | by Anders Eriksson
DTS coding
-Added progressbar for data validation
-Updated the progress bar (see figure 1)
-Fixed bug with wrong sample size (see figure 2)
-Fixed bug with unorganized barplots (see figure 2)


Exp always to the right: plotOMparams <- plotOMparams[order(plotOMparams$desc),]
plotOMparams$group <- factor(plotOMparams$group, levels=paste(unique(plotOMparams$group)))
Samplesize fix:samplesizes.annotate(boxes, as.numeric(table(plotOMparams$desc)))
Progressbar: progress <- c(round(l(100/(length(xml_list)))),round(flycount(100/(totalflies))))
Rescreening:
-Finished rescreening last Thursday. Started to evaluate the new data
Optomotor platform: Ran a few more tests to confirm that the machine was still working, it is. I also adjusted the 0 line so that it is at 0, by readjusting the “zero line” screw. Looks much better now but it is still not perfectly at 0. A difference 0.1 on the computer screen translates to 100 in the evaluation chart.
Optomotor platform:
Ran a few more tests to confirm that the 0 line is always at 0. Readjusted the “zero line” screw. Looks much better now. It is still not perfectly at 0 but a difference of 0.1 in the chart translates to 100 in the evaluation graph.
Category: Lab, Optomotor response, R code, Uncategorized | No Comments
Pooling data not possible: working on a fix
on Tuesday, June 2nd, 2020 12:25 | by Anders Eriksson
Category: R code | No Comments
Dwelling time errors, prevents code from running
on Monday, May 25th, 2020 12:28 | by Anders Eriksson
Category: R code | No Comments
Statistical evaluation of OR
on | by Anders Eriksson
Recently I measured the optomotor response in T4/T5 flies. As expected, they did not respond to the optomotor stimulus as seen in the left chart below. However, statistical evaluation struggles to quantify this difference. It might be that this is because of the low sample size, or that we are using the wrong statistical analysis?

Category: crosses, genetics, Lab, Optomotor response, R code | No Comments
Making sense of dwelling data
on Tuesday, March 24th, 2020 12:54 | by Anders Eriksson
Mean or median?
Unpunished | Punished | |
100 | 1 | |
1 | 3 | |
1,2 | 1 | |
0,5 | 2 | |
0,3 | 5 | |
5 | ||
Mean | 18 | 2,4 |
Median | 1,1 | 2 |
A highly hypothetical scenario of how the distribution of dwelling times could be. Even if unrealistic, it still illustrates the problem of using median instead of mean.
Plotting the dwelling times as means

Grouped analysis of dwelling times

Cumming estimation plot

Category: flight, Memory, Operant learning, operant self-learning, R code | No Comments
Updates on optomotor response software
on Friday, March 20th, 2020 3:19 | by Anders Eriksson
- Added a timer
- Previous version generated faulty xml files. This bug has been squashed (see image below)
- Display window of the software now also prints which direction the striped arena turns

Category: Optomotor response, R code, Uncategorized | No Comments