| title | author | date | output | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
stats |
Alex |
1/2/2022 |
|
This R notebooks prepares figures to summarize WikiPathways activity. The output files are displayed on the website and used in publications and grant applications. Please edit in coordination with the WikiPathways development team.
- Data points are collected in _data/
- Plots are saved in assets/img/
Collect history from old webservice, using getPathwayHistory on "Approved" pathways and checking oldest revision on pathways after WP3959 against a cutoff "oldest.date".
This was used to populate the data table; only run once. See GitHub Collections to update the table using data sourced from GitHub repos.
Collect monthly edit history from "User edits in month" from old server at wpi/statistics/editCounts.txt, manually simplified to editCounts.csv.
This was used to populate the data table; only run once. See GitHub Collections to update the table using data sourced from GitHub repos.
## read saved data
wpid.all.df.cnts <- read.csv("../_data/pathway_counts.csv", stringsAsFactors = F)
edits.user.df <- read.csv("../_data/edit_counts.csv", stringsAsFactors = F)
## add new row of data
# TODO: count WP folders in _pathways/
# TODO: commits per time frame, see https://git-scm.com/docs/git-rev-list with
# parms --since and --untilComposite plot for main page: pathway count and number of edits per month.
First, let's combine our data frames and make a proper date column and factor by month
combo.df <- edits.user.df %>%
full_join(wpid.all.df.cnts, by="date") %>%
dplyr::filter(!is.na(edits)) %>%
arrange(date)
combo.df$date <- strptime(paste0(combo.df$date,"01"), "%Y%m%d")
combo.df$month <- factor(format(combo.df$date, "%B"),
levels = month.name)Next, let's display the latest data points
tail(combo.df[,2:4],1)## edits pathways month
## 83 272 1991 December
Next, let's plot a time series
# RColorBrewer::display.brewer.all()
bcols <- RColorBrewer::brewer.pal(3,"Set1")
acols <- bcols
bcols <- c("#FF8120","#3955E7")
acols <- c("#D16919","#1E3199")
# date range for x-axis
Ym.end <- wpid.all.df.cnts[nrow(wpid.all.df.cnts),1]+1 #inclusive of final month
Ym.start <- Ym.end - 400 # 4 years
# scaling for primary and secondary y-axes
ylim.prim <- c(0, max(combo.df$edits, na.rm = T)) # range for edits
ylim.sec <- c(min(combo.df$pathways, na.rm = T), max(combo.df$pathways, na.rm = T)) # range for pathways
b <- diff(ylim.prim)/diff(ylim.sec)
a <- b*(ylim.prim[1] - ylim.sec[1])
p <- ggplot(combo.df) +
geom_bar(aes(x = as.Date(date),y=edits),stat="identity", fill=bcols[2]) +
geom_line(data=na.omit(combo.df),
aes(x = as.Date(date),y=a + pathways * b),
color = bcols[1], size = 1) +
scale_x_date(date_breaks = "1 year", date_labels = "%Y",
name = "",
limits = c(as.Date(strptime(paste0(Ym.start,"01"),"%Y%m%d")),as.Date(strptime(paste0(Ym.end,"01"),"%Y%m%d")))) +
scale_y_continuous(name="Edits",
limits = ylim.prim,
sec.axis=sec_axis(~ (. - a)/b,
name="Pathways")) +
ggtitle("") +
xlab("") +
theme(axis.text.y.left=element_text(colour=acols[2]),
axis.text.y.right=element_text(colour=acols[1]),
axis.ticks.y.left=element_line(colour=acols[2]),
axis.ticks.y.right=element_line(colour=acols[1]),
axis.title.y.left = element_text(colour=acols[2]),
axis.title.y.right = element_text(colour=acols[1]),
text = element_text(size = 7),
panel.grid.major = element_line(color="#eeeeee"),
panel.background = element_rect(fill='transparent'), #transparent panel bg
plot.background = element_rect(fill='transparent', color=NA), #transparent plot bg
panel.grid.minor = element_blank(), #remove minor gridlines
legend.background = element_rect(fill='transparent'), #transparent legend bg
legend.box.background = element_rect(fill='transparent') #transparent legend panel
)
p
ggsave("../assets/img/main_stats.png", plot = last_plot(),
width = 650, height = 450, units = "px", dpi = 250, bg='transparent')Now, let's make pngs per month for animation!
# plot per month
for(i in seq(nrow(combo.df),1)){
combo.df.anim<-combo.df[1:i,]
p <- ggplot(combo.df.anim) +
geom_bar(aes(x = as.Date(date),y=edits),stat="identity", fill=bcols[2]) +
geom_line(data=na.omit(combo.df.anim),
aes(x = as.Date(date),y=a + pathways * b),
color = bcols[1], size = 1) +
scale_x_date(date_breaks = "1 year", date_labels = "%Y",
name = "",
limits = c(as.Date(strptime(paste0(Ym.start,"01"),"%Y%m%d")),as.Date(strptime(paste0(Ym.end,"01"),"%Y%m%d")))) +
scale_y_continuous(name="Edits",
limits = ylim.prim,
sec.axis=sec_axis(~ (. - a)/b,
name="Pathways")) +
ggtitle("") +
xlab("") +
theme(axis.text.y.left=element_text(colour=acols[2]),
axis.text.y.right=element_text(colour=acols[1]),
axis.ticks.y.left=element_line(colour=acols[2]),
axis.ticks.y.right=element_line(colour=acols[1]),
axis.title.y.left = element_text(colour=acols[2]),
axis.title.y.right = element_text(colour=acols[1]),
text = element_text(size = 7),
panel.grid.major = element_line(color="#eeeeee"),
panel.background = element_rect(fill='transparent'), #transparent panel bg
plot.background = element_rect(fill='transparent', color=NA), #transparent plot bg
panel.grid.minor = element_blank(), #remove minor gridlines
legend.background = element_rect(fill='transparent'), #transparent legend bg
legend.box.background = element_rect(fill='transparent') #transparent legend panel
)
p
ggsave(paste0("stats_files/main_stats_",str_pad(i, 3, pad = "0"),".png"), plot = last_plot(),
width = 650, height = 450, units = "px", dpi = 250)
}## `geom_line()`: Each group consists of only one observation.
## ℹ Do you need to adjust the group aesthetic?
#make animated gif
anim.img.list <- list.files(path='stats_files', pattern = '*.png', full.names = TRUE)
anim.img.list %>%
image_read() %>% # reads each path file
image_join() %>% # joins image
image_animate(delay=as.integer(3*100/nrow(combo.df)), #first number is total seconds for all frames to play
loop = 1) %>% # number of repeat plays
image_write("../assets/img/main_stats.gif") # write to current dir
anim.img.list %>%
image_read() %>% # reads each path file
image_join() %>% # joins image
image_animate(delay=as.integer(3*100/nrow(combo.df)), #first number is total seconds for all frames to play
loop = 0) %>% # number of repeat plays
image_write("../assets/img/main_stats_inf.gif") # write to current dir
#clean up
lapply(anim.img.list, function(fn){
if (file.exists(fn))
file.remove(fn)
})