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

af and or calcs, not merging

Browse source
This commit is contained in:
Tanushree Tunstall 2020-06-18 15:57:25 +01:00
parent 010ef133dd
commit b33419c939
1 changed files with 27 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

41
scripts/AF_and_OR_calcs.R
View file

@ -9,7 +9,7 @@ setwd('~/git/LSHTM_analysis/scripts')
getwd()
options(scipen = 999) #disabling scientific notation in R.
options(scipen = 4)
#options(scipen = 4)
#%% variable assignment: input and output paths & filenames
drug = 'pyrazinamide'
@ -40,7 +40,8 @@ cat(paste0('Reading infile2: gene associated metadata:', infile_metadata))
#===========
# outdir = 'git/Data/pyrazinamide/output'
outdir = paste0('~/git/Data', '/', drug, '/', 'output')
out_filename = paste0(tolower(gene),'_', 'meta_data_with_AF_OR.csv')
#out_filename = paste0(tolower(gene), '_meta_data_with_AF_OR.csv')
out_filename = paste0(tolower(gene), '_AF_OR.csv')
outfile = paste0(outdir, '/', out_filename)
cat(paste0('Output file with full path:', outfile))
#%% end of variable assignment for input and output files
@ -261,7 +262,7 @@ statistic_chi = sapply(gene_snps_unique,function(m){
})
# statistic_chi: has suffix added of '.X-squared'
head(statistic_chi)
stat_chi = statistic_chi
# remove suffix
names(stat_chi) = gsub(".X-squared", "", names(statistic_chi))
@ -389,7 +390,6 @@ hist(log(ors)
, breaks = 100)
# sanity check: if names are equal (just for 3 vars)
all(sapply(list(names(afs)
, names(pvals_chi)
, names(statistic_chi) # should return False
@ -445,11 +445,7 @@ comb_AF_and_OR = data.frame(afs
, ci_ub_fisher
, pvals_logistic
, ors_logistic)
cat('No. of rows in comb_AF_and_OR: ', nrow(comb_AF_and_OR)
, '\nNo. of cols in comb_AF_and_OR: ', ncol(comb_AF_and_OR))
@ -464,11 +460,28 @@ if (table(rownames(comb_AF_and_OR) == comb_AF_and_OR$mutation)){
}else{
cat('FAIL: rownames and mutation col values mismatch')
}
#########################################################
# write file out: pnca_AF_OR
#########################################################
cat(paste0('writing output file: '
, '\nFilename: ', outfile))
write.csv(comb_AF_and_OR, outfile
, row.names = F)
cat(paste0('Finished writing:'
, out_filename
, '\nNo. of rows: ', nrow(comb_AF_and_OR)
, '\nNo. of cols: ', ncol(comb_AF_and_OR)))
#************************************************
cat('======================================================================')
rm(out_filename)
cat('End of script: calculated AF, OR, pvalues and saved file')
#########################################################
# 3: Merge meta data file + calculated num params
#########################################################
df1 = pnca_metadata
df1 = gene_metadata
df2 = comb_AF_and_OR
cat('checking commom col of the two dfs before merging:'
@ -509,10 +522,10 @@ if ( identical(na_count[[length(na_count)]], na_count[[length(na_count)-1]], na_
}
# reassign custom colnames
cat('Assigning custom colnames for the calculated params...')
colnames(merged_df)[colnames(merged_df)== "ors"] <- "OR"
colnames(merged_df)[colnames(merged_df)== "pvals"] <- "pvalue"
colnames(merged_df)[colnames(merged_df)== "afs"] <- "AF"
#cat('Assigning custom colnames for the calculated params...')
#colnames(merged_df)[colnames(merged_df)== "ors"] <- "OR"
#colnames(merged_df)[colnames(merged_df)== "pvals"] <- "pvalue"
#colnames(merged_df)[colnames(merged_df)== "afs"] <- "AF"
colnames(merged_df)