R scripts and utilities for analyzing Flyception2 data
Windows: R and Rtools
Mac: R and Xcode
ImageJ https://imagej.nih.gov/ij/
Set ImageJ (and Fiji if present) to save Tiff in Intel byte order from Edit -> Options -> Input/Output...
The following commands will install packages necessary for running Flyception2R.
install.packages(c("devtools", "ggplot2", "RNiftyReg", "zoo", "loggit", "installr", "Rmisc", "ggforce"))
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("EBImage")
library(devtools)
library(installr)
install.Rtools() # only for Windows
devtools::install_github("tkatsuki/dipr")
devtools::install_github("tkatsuki/Flyception2R")
library(Flyception2R)
The function Flyception2R() can process and analyze data acquired with Flyception2 semi-automatically. The minimal input needed to run Flyception2R() is the location of the directory that contains the Flyception2 video files, although you would normally want to specify which frames of the video to be analyzed, otherwise it will analyze the entire video and likely run out of memory. The following usage example demonstrates how to analyze and interpret Flyception2 data using a real 100sec recording from a male fly that expressed GCaMP6s and tdTomato in the P1 neurons during courtship.
To run the following example, first download the data (~13GB) from the link below.
https://www.dropbox.com/sh/okey9zkuhpw44zb/AABmV_mT09rPtAWEU7lgcd-Ua?dl=0
dir <- "/PATH/TO/DATA/demo_data/P1_GCaMP6s_tdTomato_06182018_CW_Dual_Laser/P1-Gal4_UAS-GCaMP6s_tdTomato_12/" # don't forget the slash at the end
Flyception2R(dir=dir, FOI=c(4242, 4556), flash=1)
Because the marker position, hence the center of the image, relative to the fly brain varies from fly to fly, the position of the window (ROI) for segmenting neurons needs to be manually adjusted. Flyception2R will ask you if the window_size and window_offset are acceptable. Check the file _redwindow.tif and adjust the window size or offset so that the neurons of your interest are in the window (see example below). When you want to move the window up, decrease the y value. When you want to move the window left, decrease the x value. You can create multiple ROIs if you wish.
[1] "Current window_size is x=68 y=28"
[1] "Current window_offset is x=-4 y=12"
Check redwindow.tif. Is the window size good (Y or N)?:Y
Check redwindow.tif. Is the window offset good (Y or N)?:N
Enter new x offset:0
Enter new y offset:0
The function spits out results like so:
window_size was x=68 y=28
window_offset was x=-4 y=12
FOI was from 4242 to 4556
Max F_ratio intensity in this bout was 0.719705128074784
Number of good frames was 197
||c(4242, 4556) ||c(68, 28) ||c(-4, 12) ||197 ||0.720 ||
Check _frgcombined_goodfr20_normalized.tif that the neurons are correctly segmented.
