agenerated corr plots with MAF and provean
This commit is contained in:
parent
1a513913ce
commit
569e372476
6 changed files with 115 additions and 104 deletions
|
|
@ -22,9 +22,12 @@
|
|||
#1) Corr type?
|
||||
#2)
|
||||
##################################################################
|
||||
# LigDist_colname #from globals: plotting_globals.R
|
||||
# ppi2Dist_colname #from globals: plotting_globals.R
|
||||
# naDist_colname #from globals: plotting_globals.R
|
||||
corr_data_extract <- function(df
|
||||
#, gene_name = gene
|
||||
, drug_name = drug
|
||||
, gene
|
||||
, drug
|
||||
#, ligand_dist_colname = LigDist_colname
|
||||
, colnames_to_extract
|
||||
, colnames_display_key
|
||||
|
|
@ -38,34 +41,48 @@ corr_data_extract <- function(df
|
|||
, "\n=========================================")
|
||||
|
||||
cat("\nExtracting default columns for"
|
||||
#, "\nGene name:", gene
|
||||
, "\nGene name:", gene
|
||||
, "\nDrug name:", drug)
|
||||
|
||||
colnames_to_extract = c(drug
|
||||
#, "mutationinformation"
|
||||
#, "mutation_info_labels"
|
||||
, "dst_mode"
|
||||
, "duet_stability_change"
|
||||
, "ligand_affinity_change"
|
||||
, "ligand_distance"
|
||||
#, ligand_dist_colname
|
||||
, "interface_dist"
|
||||
, "ddg_foldx"
|
||||
, "deepddg"
|
||||
, "asa"
|
||||
, "rsa"
|
||||
, "kd_values"
|
||||
, "rd_values"
|
||||
, "af"
|
||||
, "log10_or_mychisq"
|
||||
, "neglog_pval_fisher"
|
||||
, "ddg_dynamut2"
|
||||
, "consurf_score"
|
||||
, "snap2_score"
|
||||
, "ddg_dynamut", "ddg_encom", "dds_encom", "ddg_mcsm", "ddg_sdm", "ddg_duet"
|
||||
, "mcsm_na_affinity"
|
||||
, "mcsm_ppi2_affinity"
|
||||
)
|
||||
geneL_normal = c("pnca")
|
||||
geneL_na = c("gid", "rpob")
|
||||
geneL_ppi2 = c("alr", "embb", "katg", "rpob")
|
||||
|
||||
common_colnames = c(drug, "dst_mode"
|
||||
, "duet_stability_change" , "ddg_foldx" , "deepddg" , "ddg_dynamut2"
|
||||
, "asa" , "rsa" , "kd_values" , "rd_values"
|
||||
, "maf" , "log10_or_mychisq" , "neglog_pval_fisher"
|
||||
, LigDist_colname
|
||||
, "consurf_score" , "snap2_score" , "provean_score"
|
||||
, "ligand_affinity_change"
|
||||
#, "ddg_dynamut", "ddg_encom", "dds_encom", "ddg_mcsm", "ddg_sdm", "ddg_duet"
|
||||
)
|
||||
|
||||
display_common_colnames = c( drug, "dst_mode"
|
||||
, "DUET" , "FoldX" , "DeepDDG", "Dynamut2"
|
||||
, "ASA" , "RSA" , "KD" , "RD"
|
||||
, "MAF" , "Log(OR)" , "-Log(P)"
|
||||
, "Lig-Dist"
|
||||
, "ConSurf" , "SNAP2" , "PROVEAN"
|
||||
, "mCSM-lig"
|
||||
# , "Dynamut" , "ENCoM-DDG" , "mCSM" , "SDM" , "DUET-d" , "ENCoM-DDS"
|
||||
)
|
||||
|
||||
if (tolower(gene)%in%geneL_normal){
|
||||
colnames_to_extract = c(common_colnames)
|
||||
display_colnames = c(display_common_colnames)
|
||||
|
||||
}
|
||||
|
||||
if (tolower(gene)%in%geneL_ppi2){
|
||||
colnames_to_extract = c(common_colnames ,"mcsm_ppi2_affinity", ppi2Dist_colname)
|
||||
display_colnames = c(display_common_colnames,"mCSM-PPI2" , "PPI-Dist")
|
||||
}
|
||||
|
||||
if (tolower(gene)%in%geneL_na){
|
||||
colnames_to_extract = c(common_colnames,"mcsm_na_affinity", naDist_colname)
|
||||
display_colnames = c(display_common_colnames, "mCSM-NA", "NA-Dist")
|
||||
}
|
||||
|
||||
# [optional] arg: extract_scaled_cols
|
||||
if (extract_scaled_cols){
|
||||
|
|
@ -77,46 +94,15 @@ corr_data_extract <- function(df
|
|||
colnames_to_extract = colnames_to_extract
|
||||
}
|
||||
|
||||
corr_df = df[, colnames(df)%in%colnames_to_extract]
|
||||
|
||||
# extract df based on gene
|
||||
corr_df = df[,colnames_to_extract]
|
||||
colnames(corr_df)
|
||||
display_colnames
|
||||
|
||||
# arg: colnames_display_key
|
||||
colnames_display_key = c(duet_stability_change = "DUET"
|
||||
, ligand_affinity_change = "mCSM-lig"
|
||||
, ligand_distance = "ligand_distance"
|
||||
#, ligand_dist_colname = "ligand_distance"
|
||||
, interface_dist = "interface_dist"
|
||||
, ddg_foldx = "FoldX"
|
||||
, deepddg = "DeepDDG"
|
||||
, asa = "ASA"
|
||||
, rsa = "RSA"
|
||||
, kd_values = "KD"
|
||||
, rd_values = "RD"
|
||||
, af = "MAF"
|
||||
, log10_or_mychisq = "Log (OR)"
|
||||
, neglog_pval_fisher = "-Log (P)"
|
||||
, ddg_dynamut2 = "Dynamut2"
|
||||
, consurf_score = "Consurf"
|
||||
, snap2_score = "SNAP2"
|
||||
, ddg_dynamut = "Dynamut"
|
||||
, ddg_encom = "ENCoM-DDG"
|
||||
, ddg_mcsm = "mCSM"
|
||||
, ddg_sdm = "SDM"
|
||||
, ddg_duet = "DUET-d"
|
||||
, dds_encom = "ENCoM-DDS"
|
||||
, mcsm_na_affinity = "mCSM-NA"
|
||||
, mcsm_ppi2_affinity = "mCSM-PPI2")
|
||||
|
||||
# COMMENT: This only works when all the columns are in the namekey vector.
|
||||
# If one is missing, there is no error, but it also renamed as "NA.
|
||||
#names(corr_df) <- colnames_display_key[names(corr_df)]
|
||||
|
||||
# Solution: to use plyr::rename()
|
||||
# Consider using requireNamespace() instead of library() so its function names doesn't collide with dplyr's.
|
||||
corr_df = plyr::rename(corr_df
|
||||
, replace = colnames_display_key
|
||||
, warn_missing = T
|
||||
, warn_duplicated = T)
|
||||
|
||||
colnames(corr_df)[colnames(corr_df)%in%colnames_to_extract] <- display_colnames
|
||||
colnames(corr_df)
|
||||
|
||||
cat("\nExtracted ncols:", ncol(corr_df)
|
||||
,"\nRenaming successful")
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue