parse single mutant rows in rls.csv in R

Parse rls.csv in R to pick single-gene mutant rows.

bug on multiple entries of NAME-to-ORF mapping. This seems to be 8 bugs in file "SceORF_name.csv".

It seems that probably four ORFs share the same gene NAMES, probably due to merged annotations.  I need to revisit this annotation problem. 

The for-loop parsing of 50681 rows took about 2.5 hours on Byte, an apple laptop. 

Impressive. There 4101 single-gene RLS measures?! Visual check of ORF seems to confirm this. These covered almost all the essential genes. 

write.csv(tb,"rls_20150103.csv", quote=T, row.names=F )

#Visual check in Excel. All fields seem correct. 

R script is: "_find_LL_20150104.R", pwd ~/0.network.aging.prj/9.ken/

########  "_find_LL_20150104.R"

setwd("~/projects/0.network.aging.prj/9.ken")

rm(list=ls())

list.files(pattern='csv')

#get sce ORF and names

tb.sce = read.csv("SceORF_name.csv",header=F, colClasses = c(rep('character',2)))

str(tb.sce)

names(tb.sce)= c('ORF','NAME')

set_sce_names = unique(c(tb.sce$ORF, tb.sce$NAME)) #this is the dictionary

tb = read.csv("rls.csv", colClasses=c("integer", rep("character",8),

                                      rep("numeric",5), rep("character",8),

                                      rep("numeric",5),'character',

                                      rep('numeric',3), 'character')

              )

str(tb)

tb$setRLS_vs_ref_RLS = tb$set_lifespan_mean / tb$ref_lifespan_mean

hist(tb$setRLS_vs_ref_RLS, br = 30)

summary(tb$ranksum_p)

tb$LL = ifelse( tb$ranksum_p <0.05 & tb$setRLS_vs_ref_RLS>1, 'LL', 'NLL' )

table(tb$LL)

sort( unique(toupper(tb$set_genotype[tb$LL=='LL'])) ) 

# need to pick up single-gene mutants from tb$set_genotype 

tb$ORF = NA; tb$NAME = NA; tb$single_gene_flag = 0;

for( i in 1:length(tb[,1])) { 

#for( i in 1:100) {

  print (paste("i=",i))

  elements = strsplit(  tb$set_genotype[i], '\\s+',)

  current_name = toupper(elements[[1]][1])

  if (( current_name %in% set_sce_names) & (length(elements)==1 )){

    if (current_name %in% tb.sce$NAME) {

      tb$ORF[i] = tb.sce$ORF[tb.sce$NAME == current_name]

      tb$NAME[i] = current_name;

      tb$single_gene_flag[i] = 1;

    } else if (current_name %in% tb.sce$ORF ) {

      tb$NAME[i] = tb.sce$NAME[tb.sce$ORF == current_name]

      tb$ORF[i] = current_name;

      tb$single_gene_flag[i] = 1;  

    } else {

      print ("Excuse me. NAME or ORF seems weird at line 42. tb$single_gene_flag set to -1.")

      tb$single_gene_flag[i] = -1;

    }     

  }

} #run from 12:38- about 3pm. 

save.image("_20150104.RData")

table( tb$single_gene_flag)

length( unique(tb$ORF[tb$single_gene_flag == 1]) )

write.csv(tb,"rls_20150104.csv", quote=T, row.names=F )

######## end of  "_find_LL_20150104.R"