R/plotGCaMP_multi.R
In SenguptaLab/MF.matR: Takes mirofluidics data and does stuff with it
Defines functions
plotGCaMP_multi
Documented in
plotGCaMP_multi
#' plotGCaMP_multi
#'
#' Function is a wrapper for exp.fit.all.log.lin which outputs corrected GCaMP signals as well as plots showing
#' original and corrected signals. Also plots average trace for GCaMP signal. Inputs are matfiles with a "signal" field,
#' or .csv files from imageJ analysis.
#' The script searches recursively for matfiles (*.mat) or csv files (\*.csv) from a file (can be any placeholder file). Need to exclude
#' matfiles that are not GCaMP files either using FileFilter, or by putting only relevant files in the folder. Default
#' startpulse assumes 400ms delay for camera recording.
#' Requires max_delta helper function
#' time etc...
#' @param FileFilter string to search/subset filenames
#' @param matlab whether to use matfiles or data from ImageJ quantification. Defaults to FALSE (matfiles)
#' @param genotype label the genotype for these data
#' @param cue label the stimulus cue.
#' @param food label to food cue
#' @param startPulse begin of stimulus
#' @param endPulse endPulse time of stimulus
#' @param center_on_pulse optional parameter to center delF values by the mean of the stimulus duration
#' 'OFF' = Bring values to mean delF of 2nd half of pulse duration, order by OFF responses
#' 'ON' = Bring values to mean delF of 2nd half pre-pulse duration, order by ON responses
#' @param show.plots render plots for baseline correction - defaults to TRUE
#' @param use.Fmax normalize all amplitudes to within 0-1
#' @param neuron neuron being analyzed
#' @param linear optional argument piped into exp.fit.all.log.lin include a linear term in the fit?
#' @param heatmap_limits optional 3-value vector defining the color scale and y axis limits, ie c(-1,0,2)
#' @param folderpath user supplied path for recursive file search. If missing, it will prompt for a fiel selection.
#' @param backsub do you want to subtract background ie subtract ROI of background from the mean cell ROI? Defaults to TRUE
#' @param exp.fit do you want to use raw deltaF/F values - ie exponentially corrected? Defaults to TRUE, which means it will be corrected
#'
#' @importFrom magrittr "%>%"
#' @importFrom magrittr "%<>%"
#' @importFrom magrittr "%$%"
#' @export
#' @examples data <- plotGCaMP_multi(N2, genotype = N2, cue = octanol)
#'
plotGCaMP_multi <- function(FileFilter,
matlab = FALSE,
genotype = genotype,
cue = cue,
food = OP50,
startPulse = 29.5,
endPulse = 59.5,
center_on_pulse = "none",
show.plots = TRUE,
use.Fmax = FALSE,
neuron = GCAMP,
linear = FALSE,
nls = TRUE,
backsub = FALSE,
exp.fit = FALSE,
heatmap_limits = "auto",
folderPath,
...) {
message("this function is for analyzing calcium imaging data, for details,
go to https://github.com/mikeod38/MF.matR")
library(tidyverse)
library(magrittr)
library(patchwork)
library(scales)
FileFilter <- quo_name(enquo(FileFilter)) # make Filter usable inside other functions
genotype <- quo_name(enquo(genotype))
cue <- quo_name(enquo(cue))
food <- quo_name(enquo(food))
neuron <- quo_name(enquo(neuron))
center_on_pulse <- quo_name(enquo(center_on_pulse))
#heatmap_limits <- enquo(heatmap_limits)
if (genotype == "genotype")
{stop("error: You left out the 'genotype' argument, was this intentional?")}
if (cue == "cue")
{stop("error: You left out the 'cue' argument, was this intentional?")}
if (!center_on_pulse %in% c("none", "ON", "OFF")) {
stop("error: 'center_on_pulse' must be either 'none', 'ON'or 'OFF'")}
if(missing(folderPath)) {
folderPath <- dirname(file.choose())
}
message(paste("using files at or below the folder:", basename(folderPath)))
#### import, format and correct for photobleaching ####
if(matlab == TRUE) {
files <- list.files(file.path(folderPath), pattern = "*.mat", recursive = TRUE)
files <- files[stringr::str_detect(files, pattern = paste0(FileFilter))]
filenames <- files
files <- file.path(folderPath, files)
#df <- data.frame(x = 1, genotype = genotype, cue = cue)
data <- purrr::map(files, ~ exp.fit.all.log.lin(filename = .,
skip.time = 10,
show.plots = show.plots,
nls = nls,
startPulse = startPulse,
endPulse = ))
} else { #for matlab = FALSE
neuronfiles <- list.files(file.path(folderPath), pattern = "*neuron_results.csv", recursive = TRUE)
neuronfiles <- neuronfiles[stringr::str_detect(neuronfiles, pattern = paste0(FileFilter))]
neuronfilenames <- neuronfiles
neuronfiles <- file.path(folderPath, neuronfiles)
message("merging imageJ files")
purrr::map(
neuronfiles,
~ merge_FIJI_data(neuronfile = ., show.plots = show.plots, backsub = backsub))
files <- list.files(file.path(folderPath), pattern = "*ImageJ_data.csv", recursive = TRUE)
files <- files[stringr::str_detect(files, pattern = paste0(FileFilter))]
filenames <- files
files <- file.path(folderPath, files)
message("correcting photobleaching")
data <- purrr::map(files, ~ exp.fit.all.log.lin(
filename = .,
skip.time = 10,
show.plots = show.plots,
matlab = FALSE,
linear = linear,
nls = nls
))
}
data %<>% tibble(data = .,
animal = filenames,
animal_num = factor(seq(from = 1, to = length(filenames))),
genotype = genotype,
cue = cue,
food = food,
neuron = neuron)
if (data %>%
unnest(cols = c(data)) %>%
group_by(animal) %>%
tally() %$%
unique(n) %>%
length() != 1) stop("error: One or more of your video files are of different length, check these files")
data %>%
unnest(cols = c(data)) %>%
group_by(animal) %>%
tally() %>%
filter(n < max(n)) %>%
select(animal)
#### recenter mean values ####
if(center_on_pulse == "OFF") {
means <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
filter(time > (startPulse+endPulse)/2 & time < endPulse) %>%
summarise(mean_pulse_delF = mean(delF))
data <- left_join(data, means, by = "animal")
data %<>% unnest(cols = c(data)) %>%
group_by(animal, animal_num) %>%
mutate(delF = delF - mean_pulse_delF) %>%
nest()
}
if(center_on_pulse == "ON") {
means <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
filter(time > startPulse/2 & time < startPulse) %>%
summarise(mean_pulse_delF = mean(delF))
data <- left_join(data, means, by = "animal")
data %<>% unnest(cols = c(data)) %>%
group_by(animal,animal_num) %>%
mutate(delF = delF - mean_pulse_delF) %>%
nest()
}
#### arrange heat map settings ####
if(center_on_pulse == "OFF") {
message("centering on OFF response")
plot_order <- data %>%
unnest(cols = c(data)) %>%
group_by(animal, animal_num, .drop = TRUE) %>%
summarise(maxD = max_delta(delF, end = endPulse)) %>%
arrange(maxD)
# breaks = c(-.5,0,1.5)
# labels = c("-.5", "0", "1.5")
# limits = c(-0.5,1.5)
}
if(center_on_pulse %in% c("ON", "none")) {
message("centering on ON response")
plot_order <- data %>% unnest(cols = c(data)) %>%
group_by(animal, animal_num, .drop = TRUE) %>%
summarise(maxD = max_delta(delF, end = startPulse)) %>%
arrange(maxD)
# breaks = c(-.5,0,1.5)
# labels = c("-.5", "0", "1.5")
# limits = c(-0.5,1.5)
}
if(use.Fmax == TRUE) {
range <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
summarise(max_delF = max(delF), Fo = quantile(delF, 0.05))
data <- full_join(data, range)
data %<>% unnest(cols = c(data)) %>%
group_by(animal,animal_num) %>%
mutate(delF = (delF - Fo) / (max_delF - Fo)) %>%
nest()
# breaks = c(0,0.5,1)
# labels = c("0", "0.5", "1")
# limits = c(0,1)
}
if(!is.numeric(heatmap_limits)) { # using auto calc unless a numeric vector input
breaks <- round(
data %>% unnest(cols = c(data)) %$% quantile(delF, c(0.05, 0.5, 0.99)),
2
)
labels <- as.character(breaks)
limits <- breaks[c(1,3)]
} else {
breaks <- heatmap_limits
labels <- as.character(breaks)
limits <- breaks[c(1,3)]
}
if(exp.fit == FALSE) {
plot1 <- data %>% unnest(cols = c(data)) %>%
ggplot(aes(x = time, y = delF)) +
geom_line(aes(group = animal), alpha = 0.1) +
geom_smooth(method = "loess", span = 0.05) +
theme_classic() +
geom_segment(aes(x = !!startPulse,
y = limits[2],
xend = !!endPulse,
yend = limits[2])) +
annotate(geom = "text",
label = cue,
y = 1.2*limits[2],
x = (startPulse + endPulse)/2,
size = 6) +
annotate(geom = "text",
label = genotype,
y = 1.2*limits[2],
x = 10,
fontface = "italic",
size = 6) +
coord_cartesian(ylim = c(limits[1], 1.5*limits[2])) +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.title.x = element_blank())
plot2 <- full_join(data, plot_order) %>%
unnest(cols = c(data)) %>%
group_by(animal_num) %>%
ggplot(aes(x = time, y = fct_reorder(animal_num, maxD))) +
geom_tile(aes(fill = delF)) +
scale_fill_viridis_c(option = "magma",
breaks = breaks,
labels = labels,
limits = limits,
oob =squish) +
theme_classic() +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.text.y = element_blank()) +
labs(y = "Animal number")
} else {
plot1 <- data %>% unnest(cols = c(data)) %>%
ggplot(aes(x = time, y = signal)) +
geom_line(aes(group = animal), alpha = 0.1) +
geom_smooth(method = "loess", span = 0.05) +
theme_classic() +
geom_segment(aes(x = !!startPulse,
y = limits[2],
xend = !!endPulse,
yend = limits[2])) +
annotate(geom = "text",
label = cue,
y = 1.2*limits[2],
x = (startPulse + endPulse)/2,
size = 6) +
annotate(geom = "text",
label = genotype,
y = 1.2*limits[2],
x = 10,
fontface = "italic",
size = 6) +
coord_cartesian(ylim = c(limits[1], 1.5*limits[2])) +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.title.x = element_blank())
plot2 <- full_join(data, plot_order) %>%
unnest(cols = c(data)) %>%
group_by(animal_num) %>%
ggplot(aes(x = time, y = fct_reorder(animal_num, maxD))) +
geom_tile(aes(fill = signal)) +
scale_fill_viridis_c(option = "magma",
breaks = breaks,
labels = labels,
limits = limits,
oob =squish) +
theme_classic() +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.text.y = element_blank()) +
labs(y = "Animal number")
}
plots <- plot1 + plot2 + plot_layout(ncol = 1, heights = c(2,1))
if(use.Fmax == TRUE) {
ggsave(plots, filename = file.path(folderPath,paste0(genotype,"_",cue,neuron,"_delFmaxplots.png")),
width = 11, height = 8.5, units = "in")
} else {
ggsave(plots, filename = file.path(folderPath,paste0(genotype,"_",cue,"_",neuron,"_",food,"_plots.png")),
width = 11, height = 8.5, units = "in")
}
write_csv(unnest(data, cols = c(data)), file = file.path(folderPath,paste0(genotype,"_",cue,"_",neuron,"_",food,"_rawdata.csv")))
return(list(data = dplyr::full_join(data, plot_order), plot = plots))
}
SenguptaLab/MF.matR documentation built on Feb. 5, 2023, 4:57 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions plotGCaMP_multi
Documented in plotGCaMP_multi
#' plotGCaMP_multi
#'
#' Function is a wrapper for exp.fit.all.log.lin which outputs corrected GCaMP signals as well as plots showing
#' original and corrected signals. Also plots average trace for GCaMP signal. Inputs are matfiles with a "signal" field,
#' or .csv files from imageJ analysis.
#' The script searches recursively for matfiles (*.mat) or csv files (\*.csv) from a file (can be any placeholder file). Need to exclude
#' matfiles that are not GCaMP files either using FileFilter, or by putting only relevant files in the folder. Default
#' startpulse assumes 400ms delay for camera recording.
#' Requires max_delta helper function
#' time etc...
#' @param FileFilter string to search/subset filenames
#' @param matlab whether to use matfiles or data from ImageJ quantification. Defaults to FALSE (matfiles)
#' @param genotype label the genotype for these data
#' @param cue label the stimulus cue.
#' @param food label to food cue
#' @param startPulse begin of stimulus
#' @param endPulse endPulse time of stimulus
#' @param center_on_pulse optional parameter to center delF values by the mean of the stimulus duration
#' 'OFF' = Bring values to mean delF of 2nd half of pulse duration, order by OFF responses
#' 'ON' = Bring values to mean delF of 2nd half pre-pulse duration, order by ON responses
#' @param show.plots render plots for baseline correction - defaults to TRUE
#' @param use.Fmax normalize all amplitudes to within 0-1
#' @param neuron neuron being analyzed
#' @param linear optional argument piped into exp.fit.all.log.lin include a linear term in the fit?
#' @param heatmap_limits optional 3-value vector defining the color scale and y axis limits, ie c(-1,0,2)
#' @param folderpath user supplied path for recursive file search. If missing, it will prompt for a fiel selection.
#' @param backsub do you want to subtract background ie subtract ROI of background from the mean cell ROI? Defaults to TRUE
#' @param exp.fit do you want to use raw deltaF/F values - ie exponentially corrected? Defaults to TRUE, which means it will be corrected
#'
#' @importFrom magrittr "%>%"
#' @importFrom magrittr "%<>%"
#' @importFrom magrittr "%$%"
#' @export
#' @examples data <- plotGCaMP_multi(N2, genotype = N2, cue = octanol)
#'
plotGCaMP_multi <- function(FileFilter,
matlab = FALSE,
genotype = genotype,
cue = cue,
food = OP50,
startPulse = 29.5,
endPulse = 59.5,
center_on_pulse = "none",
show.plots = TRUE,
use.Fmax = FALSE,
neuron = GCAMP,
linear = FALSE,
nls = TRUE,
backsub = FALSE,
exp.fit = FALSE,
heatmap_limits = "auto",
folderPath,
...) {
message("this function is for analyzing calcium imaging data, for details,
go to https://github.com/mikeod38/MF.matR")
library(tidyverse)
library(magrittr)
library(patchwork)
library(scales)
FileFilter <- quo_name(enquo(FileFilter)) # make Filter usable inside other functions
genotype <- quo_name(enquo(genotype))
cue <- quo_name(enquo(cue))
food <- quo_name(enquo(food))
neuron <- quo_name(enquo(neuron))
center_on_pulse <- quo_name(enquo(center_on_pulse))
#heatmap_limits <- enquo(heatmap_limits)
if (genotype == "genotype")
{stop("error: You left out the 'genotype' argument, was this intentional?")}
if (cue == "cue")
{stop("error: You left out the 'cue' argument, was this intentional?")}
if (!center_on_pulse %in% c("none", "ON", "OFF")) {
stop("error: 'center_on_pulse' must be either 'none', 'ON'or 'OFF'")}
if(missing(folderPath)) {
folderPath <- dirname(file.choose())
}
message(paste("using files at or below the folder:", basename(folderPath)))
#### import, format and correct for photobleaching ####
if(matlab == TRUE) {
files <- list.files(file.path(folderPath), pattern = "*.mat", recursive = TRUE)
files <- files[stringr::str_detect(files, pattern = paste0(FileFilter))]
filenames <- files
files <- file.path(folderPath, files)
#df <- data.frame(x = 1, genotype = genotype, cue = cue)
data <- purrr::map(files, ~ exp.fit.all.log.lin(filename = .,
skip.time = 10,
show.plots = show.plots,
nls = nls,
startPulse = startPulse,
endPulse = ))
} else { #for matlab = FALSE
neuronfiles <- list.files(file.path(folderPath), pattern = "*neuron_results.csv", recursive = TRUE)
neuronfiles <- neuronfiles[stringr::str_detect(neuronfiles, pattern = paste0(FileFilter))]
neuronfilenames <- neuronfiles
neuronfiles <- file.path(folderPath, neuronfiles)
message("merging imageJ files")
purrr::map(
neuronfiles,
~ merge_FIJI_data(neuronfile = ., show.plots = show.plots, backsub = backsub))
files <- list.files(file.path(folderPath), pattern = "*ImageJ_data.csv", recursive = TRUE)
files <- files[stringr::str_detect(files, pattern = paste0(FileFilter))]
filenames <- files
files <- file.path(folderPath, files)
message("correcting photobleaching")
data <- purrr::map(files, ~ exp.fit.all.log.lin(
filename = .,
skip.time = 10,
show.plots = show.plots,
matlab = FALSE,
linear = linear,
nls = nls
))
}
data %<>% tibble(data = .,
animal = filenames,
animal_num = factor(seq(from = 1, to = length(filenames))),
genotype = genotype,
cue = cue,
food = food,
neuron = neuron)
if (data %>%
unnest(cols = c(data)) %>%
group_by(animal) %>%
tally() %$%
unique(n) %>%
length() != 1) stop("error: One or more of your video files are of different length, check these files")
data %>%
unnest(cols = c(data)) %>%
group_by(animal) %>%
tally() %>%
filter(n < max(n)) %>%
select(animal)
#### recenter mean values ####
if(center_on_pulse == "OFF") {
means <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
filter(time > (startPulse+endPulse)/2 & time < endPulse) %>%
summarise(mean_pulse_delF = mean(delF))
data <- left_join(data, means, by = "animal")
data %<>% unnest(cols = c(data)) %>%
group_by(animal, animal_num) %>%
mutate(delF = delF - mean_pulse_delF) %>%
nest()
}
if(center_on_pulse == "ON") {
means <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
filter(time > startPulse/2 & time < startPulse) %>%
summarise(mean_pulse_delF = mean(delF))
data <- left_join(data, means, by = "animal")
data %<>% unnest(cols = c(data)) %>%
group_by(animal,animal_num) %>%
mutate(delF = delF - mean_pulse_delF) %>%
nest()
}
#### arrange heat map settings ####
if(center_on_pulse == "OFF") {
message("centering on OFF response")
plot_order <- data %>%
unnest(cols = c(data)) %>%
group_by(animal, animal_num, .drop = TRUE) %>%
summarise(maxD = max_delta(delF, end = endPulse)) %>%
arrange(maxD)
# breaks = c(-.5,0,1.5)
# labels = c("-.5", "0", "1.5")
# limits = c(-0.5,1.5)
}
if(center_on_pulse %in% c("ON", "none")) {
message("centering on ON response")
plot_order <- data %>% unnest(cols = c(data)) %>%
group_by(animal, animal_num, .drop = TRUE) %>%
summarise(maxD = max_delta(delF, end = startPulse)) %>%
arrange(maxD)
# breaks = c(-.5,0,1.5)
# labels = c("-.5", "0", "1.5")
# limits = c(-0.5,1.5)
}
if(use.Fmax == TRUE) {
range <- data %>% unnest(cols = c(data)) %>%
group_by(animal) %>%
summarise(max_delF = max(delF), Fo = quantile(delF, 0.05))
data <- full_join(data, range)
data %<>% unnest(cols = c(data)) %>%
group_by(animal,animal_num) %>%
mutate(delF = (delF - Fo) / (max_delF - Fo)) %>%
nest()
# breaks = c(0,0.5,1)
# labels = c("0", "0.5", "1")
# limits = c(0,1)
}
if(!is.numeric(heatmap_limits)) { # using auto calc unless a numeric vector input
breaks <- round(
data %>% unnest(cols = c(data)) %$% quantile(delF, c(0.05, 0.5, 0.99)),
2
)
labels <- as.character(breaks)
limits <- breaks[c(1,3)]
} else {
breaks <- heatmap_limits
labels <- as.character(breaks)
limits <- breaks[c(1,3)]
}
if(exp.fit == FALSE) {
plot1 <- data %>% unnest(cols = c(data)) %>%
ggplot(aes(x = time, y = delF)) +
geom_line(aes(group = animal), alpha = 0.1) +
geom_smooth(method = "loess", span = 0.05) +
theme_classic() +
geom_segment(aes(x = !!startPulse,
y = limits[2],
xend = !!endPulse,
yend = limits[2])) +
annotate(geom = "text",
label = cue,
y = 1.2*limits[2],
x = (startPulse + endPulse)/2,
size = 6) +
annotate(geom = "text",
label = genotype,
y = 1.2*limits[2],
x = 10,
fontface = "italic",
size = 6) +
coord_cartesian(ylim = c(limits[1], 1.5*limits[2])) +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.title.x = element_blank())
plot2 <- full_join(data, plot_order) %>%
unnest(cols = c(data)) %>%
group_by(animal_num) %>%
ggplot(aes(x = time, y = fct_reorder(animal_num, maxD))) +
geom_tile(aes(fill = delF)) +
scale_fill_viridis_c(option = "magma",
breaks = breaks,
labels = labels,
limits = limits,
oob =squish) +
theme_classic() +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.text.y = element_blank()) +
labs(y = "Animal number")
} else {
plot1 <- data %>% unnest(cols = c(data)) %>%
ggplot(aes(x = time, y = signal)) +
geom_line(aes(group = animal), alpha = 0.1) +
geom_smooth(method = "loess", span = 0.05) +
theme_classic() +
geom_segment(aes(x = !!startPulse,
y = limits[2],
xend = !!endPulse,
yend = limits[2])) +
annotate(geom = "text",
label = cue,
y = 1.2*limits[2],
x = (startPulse + endPulse)/2,
size = 6) +
annotate(geom = "text",
label = genotype,
y = 1.2*limits[2],
x = 10,
fontface = "italic",
size = 6) +
coord_cartesian(ylim = c(limits[1], 1.5*limits[2])) +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.title.x = element_blank())
plot2 <- full_join(data, plot_order) %>%
unnest(cols = c(data)) %>%
group_by(animal_num) %>%
ggplot(aes(x = time, y = fct_reorder(animal_num, maxD))) +
geom_tile(aes(fill = signal)) +
scale_fill_viridis_c(option = "magma",
breaks = breaks,
labels = labels,
limits = limits,
oob =squish) +
theme_classic() +
theme(axis.text = element_text(size = 16),
axis.title = element_text(size = 18),
axis.text.y = element_blank()) +
labs(y = "Animal number")
}
plots <- plot1 + plot2 + plot_layout(ncol = 1, heights = c(2,1))
if(use.Fmax == TRUE) {
ggsave(plots, filename = file.path(folderPath,paste0(genotype,"_",cue,neuron,"_delFmaxplots.png")),
width = 11, height = 8.5, units = "in")
} else {
ggsave(plots, filename = file.path(folderPath,paste0(genotype,"_",cue,"_",neuron,"_",food,"_plots.png")),
width = 11, height = 8.5, units = "in")
}
write_csv(unnest(data, cols = c(data)), file = file.path(folderPath,paste0(genotype,"_",cue,"_",neuron,"_",food,"_rawdata.csv")))
return(list(data = dplyr::full_join(data, plot_order), plot = plots))
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.