tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

a massive waste of time

Browse source
This commit is contained in:
Tanushree Tunstall 2022-08-22 13:05:53 +01:00
parent 8d6c148fff
commit 4147a6b90f
3 changed files with 726 additions and 620 deletions
Download patch file Download diff file Expand all files Collapse all files

61
scripts/functions/dm_om_data.R
View file

@ -8,7 +8,7 @@
##################################################################
# from plotting_globals.R
# DistCutOff, LigDist_colname, ppi2Dist_colname, naDist_colname
gene
#gene
dm_om_wf_lf_data <- function(df
, gene # from globals
@ -28,9 +28,15 @@ dm_om_wf_lf_data <- function(df
sum(is.na(df$maf2))
# Initialise the required dfs based on gene name
#geneL_normal = c("pnca")
#geneL_na = c("gid", "rpob")
#geneL_ppi2 = c("alr", "embb", "katg", "rpob")
#ADDED: IMPORTANT for rpob to be in both to make sure all data is returned
geneL_normal = c("pnca")
geneL_na = c("gid", "rpob")
geneL_ppi2 = c("alr", "embb", "katg", "rpob")
geneL_both = c("rpob")
geneL_ppi2 = c("alr", "embb", "katg")
geneL_na = c("gid")
# common_dfs
common_dfsL = list(
@ -59,6 +65,14 @@ dm_om_wf_lf_data <- function(df
wf_lf_dataL = common_dfsL
}
if (tolower(gene)%in%geneL_ppi2){
additional_dfL = list(
wf_mcsm_ppi2 = data.frame()
, lf_mcsm_ppi2 = data.frame()
)
wf_lf_dataL = c(common_dfsL, additional_dfL)
}
if (tolower(gene)%in%geneL_na){
additional_dfL = list(
wf_mcsm_na = data.frame()
@ -67,13 +81,16 @@ dm_om_wf_lf_data <- function(df
wf_lf_dataL = c(common_dfsL, additional_dfL)
}
if (tolower(gene)%in%geneL_ppi2){
if (tolower(gene)%in%geneL_both){
additional_dfL = list(
wf_mcsm_ppi2 = data.frame()
, lf_mcsm_ppi2 = data.frame()
wf_mcsm_ppi2 = data.frame(),
lf_mcsm_ppi2 = data.frame(),
wf_mcsm_na = data.frame(),
lf_mcsm_na = data.frame()
)
wf_lf_dataL = c(common_dfsL, additional_dfL)
}
cat("\nInitializing an empty list of length:"
, length(wf_lf_dataL))
@ -237,6 +254,38 @@ dm_om_wf_lf_data <- function(df
}
if (tolower(gene)%in%geneL_both){
colnames_to_extract = c(
common_colnames,
"mcsm_ppi2_affinity" ,
"mcsm_ppi2_scaled" ,
"mcsm_ppi2_outcome" ,
ppi2Dist_colname,
"mcsm_na_affinity" ,
"mcsm_na_scaled" ,
"mcsm_na_outcome" ,
naDist_colname
)
display_colnames = c(
display_common_colnames,
"mcsm_ppi2_affinity",
mcsm_ppi2_dn,
"mcsm_ppi2_outcome",
ppi2_dist_dn,
"mcsm_na_affinity",
mcsm_na_dn,
"mcsm_na_outcome",
na_dist_dn
)
comb_df_sl = df[, colnames_to_extract]
colnames(comb_df_sl) = display_colnames
comb_df_sl_ppi2 = comb_df_sl[comb_df_sl[[ppi2_dist_dn]]<DistCutOff,]
comb_df_sl_na = comb_df_sl[comb_df_sl[[na_dist_dn]]<DistCutOff,]
static_cols_end = c(na_dist_dn, static_cols_end_common)
}
# Affinity filtered data: mcsm-lig: COMMON for all genes, mcsm-lig --> LigDist_colname
comb_df_sl_lig = comb_df_sl[comb_df_sl[[lig_dist_dn]]<DistCutOff,]

52
scripts/functions/plotting_data.R
View file

@ -12,7 +12,6 @@ geneL_na = c("gid", "rpob")
geneL_ppi2 = c("alr", "embb", "katg", "rpob")
if (tolower(gene)%in%geneL_na){
infilename_nca = paste0("/home/tanu/git/Misc/mcsm_na_dist/"
, tolower(gene), "_nca_distances.csv")
}
@ -72,25 +71,65 @@ cat("\nNo. of unique mutational positions:"); cat(length(upos), "\n")
#===============================================
# ADD : na distance column for genes with nucleic acid affinity
#===============================================
#gid_na_distcol
if (tolower(gene)%in%geneL_na){
# if (tolower(gene)%in%geneL_na){
#
# distcol_nca_name = read.csv(infilename_nca, header = F)
# head(distcol_nca_name)
# colnames(distcol_nca_name) <- c("mutationinformation", "nca_distance")
# head(distcol_nca_name)
# class(distcol_nca_name)
#
# mcol = colnames(distcol_nca_name)[colnames(distcol_nca_name)%in%colnames(my_df_u)]
# mcol
# head(my_df_u$mutationinformation)
# head(distcol_nca_name$mutationinformation)
#
# my_df_u = merge(my_df_u, distcol_nca_name,
# by = "mutationinformation",
# all = T)
#
# }
if (tolower(gene)%in%geneL_na){
distcol_nca_name = read.csv(infilename_nca, header = F)
if (tolower(gene)=='rpob'){
print('WARNING: running special-case handler for rpoB')
# create 5uhc equivalent column for mutationinformation
my_df_u$X5uhc_mutationinformation = paste0(my_df_u$wild_type,
my_df_u$X5uhc_position,
my_df_u$mutant_type)
colnames(distcol_nca_name) <- c("X5uhc_mutationinformation", "nca_distance")
# do stuff here
mcol = colnames(distcol_nca_name)[colnames(distcol_nca_name)%in%colnames(my_df_u)]
cat(paste0("\nMerging for gene: ", tolower(gene), "\non column: ", mcol))
head(my_df_u$mutationinformation)
head(distcol_nca_name$X5uhc_mutationinformation)
my_df_u = merge(my_df_u, distcol_nca_name,
by = "X5uhc_mutationinformation",
all = T)
} else {
head(distcol_nca_name)
colnames(distcol_nca_name) <- c("mutationinformation", "nca_distance")
head(distcol_nca_name)
class(distcol_nca_name)
mcol = colnames(distcol_nca_name)[colnames(distcol_nca_name)%in%colnames(my_df_u)]
mcol
cat(paste0("\nMerging for gene: ", tolower(gene), "\non column: ", mcol))
head(my_df_u$mutationinformation)
head(distcol_nca_name$mutationinformation)
my_df_u = merge(my_df_u, distcol_nca_name,
by = "mutationinformation",
all = T)
}
}
#===============================================
# extract mutations <10 Angstroms and symbol
#===============================================
@ -111,3 +150,4 @@ return(all_df)
########################################################################
# end of data extraction and cleaning for plots #
########################################################################

29
scripts/plotting/get_plotting_dfs.R
View file

@ -60,8 +60,8 @@ pd_df = plotting_data(mcsm_df
my_df = pd_df[[1]]
my_df_u = pd_df[[2]] # this forms one of the input for combining_dfs_plotting()
max_ang <- round(max(my_df_u[LigDist_colname]))
min_ang <- round(min(my_df_u[LigDist_colname]))
max_ang <- round(max(my_df_u[[LigDist_colname]]))
min_ang <- round(min(my_df_u[[LigDist_colname]]))
cat("\nLigand distance colname:", LigDist_colname
, "\nThe max distance", gene, "structure df" , ":", max_ang, "\u212b"
@ -128,6 +128,11 @@ geneL_normal = c("pnca")
geneL_na = c("gid", "rpob")
geneL_ppi2 = c("alr", "embb", "katg", "rpob")
# geneL_normal = c("pnca")
# geneL_both = c("rpob")
# geneL_ppi2 = c("alr", "embb", "katg")
# geneL_na = c("gid")
all_dm_om_df = dm_om_wf_lf_data(df = merged_df3, gene = gene)
wf_duet = all_dm_om_df[['wf_duet']]
@ -158,15 +163,27 @@ lf_provean = all_dm_om_df[['lf_provean']]
wf_dist_gen = all_dm_om_df[['wf_dist_gen']]
lf_dist_gen = all_dm_om_df[['lf_dist_gen']]
# ppi2 genes
if (tolower(gene)%in%geneL_ppi2){
wf_mcsm_ppi2 = all_dm_om_df[['wf_mcsm_ppi2']]
lf_mcsm_ppi2 = all_dm_om_df[['lf_mcsm_ppi2']]
}
# na genes
if (tolower(gene)%in%geneL_na){
wf_mcsm_na = all_dm_om_df[['wf_mcsm_na']]
lf_mcsm_na = all_dm_om_df[['lf_mcsm_na']]
}
if (tolower(gene)%in%geneL_ppi2){
wf_mcsm_ppi2 = all_dm_om_df[['wf_mcsm_ppi2']]
lf_mcsm_ppi2 = all_dm_om_df[['lf_mcsm_ppi2']]
}
# both ppi2+na genes:: NOT NEEDED Here as its is handled by the two ifs above
# if (tolower(gene)%in%geneL_both){
# wf_mcsm_ppi2 = all_dm_om_df[['wf_mcsm_ppi2']]
# lf_mcsm_ppi2 = all_dm_om_df[['lf_mcsm_ppi2']]
#
# wf_mcsm_na = all_dm_om_df[['wf_mcsm_na']]
# lf_mcsm_na = all_dm_om_df[['lf_mcsm_na']]
# }
s2 = c("\nSuccessfully sourced other_plots_data.R")
cat(s2)