graphs for PS lineage dist for all and dr muts
This commit is contained in:
parent
3c20be5615
commit
4de4549430
4 changed files with 93 additions and 567 deletions
|
|
@ -1,5 +1,5 @@
|
|||
getwd()
|
||||
setwd("~/git/LSHTM_analysis/mcsm_analysis/pyrazinamide/scripts/plotting") # thinkpad
|
||||
setwd("~/git/LSHTM_analysis/mcsm_analysis/pyrazinamide/scripts/plotting")
|
||||
getwd()
|
||||
|
||||
########################################################################
|
||||
|
|
@ -24,11 +24,11 @@ source("../combining_two_df.R")
|
|||
#==========================
|
||||
# This will return:
|
||||
|
||||
# df with NA:
|
||||
# df with NA for pyrazinamide:
|
||||
# merged_df2
|
||||
# merged_df3
|
||||
|
||||
# df without NA:
|
||||
# df without NA for pyrazinamide:
|
||||
# merged_df2_comp
|
||||
# merged_df3_comp
|
||||
#===========================
|
||||
|
|
@ -38,14 +38,17 @@ source("../combining_two_df.R")
|
|||
# you need merged_df2 or merged_df2_comp
|
||||
# since this is one-many relationship
|
||||
# i.e the same SNP can belong to multiple lineages
|
||||
# using the _comp dataset means
|
||||
# we lose some muts and at this level, we should use
|
||||
# as much info as available, hence use df with NA
|
||||
###########################
|
||||
|
||||
# uncomment as necessary
|
||||
#<<<<<<<<<<<<<<<<<<<<<<<<<
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!
|
||||
# REASSIGNMENT
|
||||
my_df = merged_df2
|
||||
#my_df = merged_df2_comp
|
||||
#<<<<<<<<<<<<<<<<<<<<<<<<<
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!
|
||||
|
||||
# delete variables not required
|
||||
rm(merged_df2, merged_df2_comp, merged_df3, merged_df3_comp)
|
||||
|
|
@ -59,12 +62,39 @@ is.factor(my_df$lineage)
|
|||
my_df$lineage = as.factor(my_df$lineage)
|
||||
is.factor(my_df$lineage)
|
||||
|
||||
table(my_df$mutation_info)
|
||||
table(my_df$mutation_info); str(my_df$mutation_info)
|
||||
|
||||
# subset df with dr muts only
|
||||
my_df_dr = subset(my_df, mutation_info == "dr_mutations_pyrazinamide")
|
||||
|
||||
########################################################################
|
||||
# end of data extraction and cleaning for plots #
|
||||
########################################################################
|
||||
|
||||
#==========================
|
||||
# Data for plot: assign as
|
||||
# necessary
|
||||
#===========================
|
||||
|
||||
# uncomment as necessary
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!
|
||||
# REASSIGNMENT
|
||||
|
||||
#==================
|
||||
# data for ALL muts
|
||||
#==================
|
||||
plot_df = my_df
|
||||
my_plot_name = 'lineage_dist_PS.svg'
|
||||
#my_plot_name = 'lineage_dist_PS_comp.svg'
|
||||
|
||||
#=======================
|
||||
# data for dr_muts ONLY
|
||||
#=======================
|
||||
#plot_df = my_df_dr
|
||||
#my_plot_name = 'lineage_dist_dr_PS.svg'
|
||||
#my_plot_name = 'lineage_dist_dr_PS_comp.svg'
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!
|
||||
|
||||
#==========================
|
||||
# Plot: Lineage Distribution
|
||||
# x = mcsm_values, y = dist
|
||||
|
|
@ -74,6 +104,7 @@ table(my_df$mutation_info)
|
|||
#===================
|
||||
# Data for plots
|
||||
#===================
|
||||
table(plot_df$lineage); str(plot_df$lineage)
|
||||
|
||||
# subset only lineages1-4
|
||||
sel_lineages = c("lineage1"
|
||||
|
|
@ -82,34 +113,29 @@ sel_lineages = c("lineage1"
|
|||
, "lineage4")
|
||||
|
||||
# uncomment as necessary
|
||||
df_lin = subset(my_df, subset = lineage %in% sel_lineages )
|
||||
df_lin = subset(plot_df, subset = lineage %in% sel_lineages )
|
||||
|
||||
# refactor
|
||||
df_lin$lineage = factor(df_lin$lineage)
|
||||
|
||||
table(df_lin$lineage) #{RESULT: No of samples within lineage}
|
||||
#lineage1 lineage2 lineage3 lineage4
|
||||
#104 1293 264 1311
|
||||
|
||||
# when merged_df2_comp is used
|
||||
#lineage1 lineage2 lineage3 lineage4
|
||||
#99 1275 263 1255
|
||||
|
||||
length(unique(df_lin$Mutationinformation))
|
||||
#{Result: No. of unique mutations the 4 lineages contribute to}
|
||||
|
||||
# sanity checks
|
||||
r1 = 2:5 # when merged_df2 used: because there is missing lineages
|
||||
if(sum(table(my_df$lineage)[r1]) == nrow(df_lin)) {
|
||||
if(sum(table(plot_df$lineage)[r1]) == nrow(df_lin)) {
|
||||
print ("sanity check passed: numbers match")
|
||||
} else{
|
||||
print("Error!: check your numbers")
|
||||
}
|
||||
|
||||
#<<<<<<<<<<<<<<<<<<<<<<<<<
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!!!
|
||||
# REASSIGNMENT
|
||||
df <- df_lin
|
||||
#<<<<<<<<<<<<<<<<<<<<<<<<<
|
||||
#!!!!!!!!!!!!!!!!!!!!!!!!!
|
||||
|
||||
rm(df_lin)
|
||||
|
||||
|
|
@ -117,8 +143,8 @@ rm(df_lin)
|
|||
# generate distribution plot of lineages
|
||||
#******************
|
||||
# basic: could improve this!
|
||||
library(plotly)
|
||||
library(ggridges)
|
||||
#library(plotly)
|
||||
#library(ggridges)
|
||||
|
||||
g <- ggplot(df, aes(x = ratioDUET)) +
|
||||
geom_density(aes(fill = DUET_outcome)
|
||||
|
|
@ -129,64 +155,68 @@ g <- ggplot(df, aes(x = ratioDUET)) +
|
|||
ggplotly(g)
|
||||
|
||||
# 2 : ggridges (good!)
|
||||
|
||||
my_ats = 15 # axis text size
|
||||
my_als = 20 # axis label size
|
||||
|
||||
fooNames=c('Lineage 1', 'Lineage 2', 'Lineage 3', 'Lineage 4')
|
||||
names(fooNames)=c('lineage1', 'lineage2', 'lineage3', 'lineage4')
|
||||
my_labels = c('Lineage 1', 'Lineage 2', 'Lineage 3', 'Lineage 4')
|
||||
names(my_labels) = c('lineage1', 'lineage2', 'lineage3', 'lineage4')
|
||||
|
||||
# set output dir for plots
|
||||
getwd()
|
||||
setwd("~/git/Data/pyrazinamide/output/plots")
|
||||
getwd()
|
||||
|
||||
svg('lineage_dist_PS.svg')
|
||||
# check plot name
|
||||
my_plot_name
|
||||
|
||||
printFile = ggplot( df, aes(x = ratioDUET
|
||||
, y = DUET_outcome) )+
|
||||
# output svg
|
||||
svg(my_plot_name)
|
||||
printFile = ggplot(df, aes(x = ratioDUET
|
||||
, y = DUET_outcome))+
|
||||
|
||||
#printFile=geom_density_ridges_gradient(
|
||||
geom_density_ridges_gradient( aes(fill = ..x..)
|
||||
geom_density_ridges_gradient(aes(fill = ..x..)
|
||||
, scale = 3
|
||||
, size = 0.3 ) +
|
||||
facet_wrap( ~lineage
|
||||
, scales = "free"
|
||||
# , switch = 'x'
|
||||
, labeller = labeller(lineage = fooNames) ) +
|
||||
, labeller = labeller(lineage = my_labels) ) +
|
||||
coord_cartesian( xlim = c(-1, 1)
|
||||
# , ylim = c(0, 6)
|
||||
# , clip = "off"
|
||||
) +
|
||||
scale_fill_gradientn( colours = c("#f8766d", "white", "#00bfc4")
|
||||
# , ylim = c(0, 6)
|
||||
# , clip = "off"
|
||||
) +
|
||||
scale_fill_gradientn(colours = c("#f8766d", "white", "#00bfc4")
|
||||
, name = "DUET" ) +
|
||||
theme( axis.text.x = element_text( size = my_ats
|
||||
theme(axis.text.x = element_text(size = my_ats
|
||||
, angle = 90
|
||||
, hjust = 1
|
||||
, vjust = 0.4)
|
||||
# , axis.text.y = element_text( size = my_ats
|
||||
# , angle = 0
|
||||
# , hjust = 1
|
||||
# , vjust = 0)
|
||||
# , axis.text.y = element_text(size = my_ats
|
||||
# , angle = 0
|
||||
# , hjust = 1
|
||||
# , vjust = 0)
|
||||
, axis.text.y = element_blank()
|
||||
, axis.title.x = element_blank()
|
||||
, axis.title.y = element_blank()
|
||||
, axis.ticks.y = element_blank()
|
||||
, plot.title = element_blank()
|
||||
, strip.text = element_text(size=my_als)
|
||||
, legend.text = element_text(size=10)
|
||||
, legend.title = element_text(size=my_als)
|
||||
# , legend.position = c(0.3, 0.8)
|
||||
# , legend.key.height = unit(1, 'mm')
|
||||
)
|
||||
, strip.text = element_text(size = my_als)
|
||||
, legend.text = element_text(size = 10)
|
||||
, legend.title = element_text(size = my_als)
|
||||
# , legend.position = c(0.3, 0.8)
|
||||
# , legend.key.height = unit(1, 'mm')
|
||||
)
|
||||
|
||||
print(printFile)
|
||||
dev.off()
|
||||
|
||||
#=!=!=!=!=!=!
|
||||
# COMMENT: When you look at all mutations, the lineage differences disappear...
|
||||
#=!=!=!=!=!=!=!
|
||||
# COMMENT: Not much differences in the distributions
|
||||
# when using merged_df2 or merged_df2_comp.
|
||||
# Also, the lineage differences disappear when looking at all muts
|
||||
# The pattern we are interested in is possibly only for dr_mutations
|
||||
#=!=!=!=!=!=!
|
||||
#=!=!=!=!=!=!=!
|
||||
#===================================================
|
||||
|
||||
# COMPARING DISTRIBUTIONS
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue