#!/usr/bin/Rscript   
getwd()
setwd('~/git/mosaic_2020/')
getwd()
########################################################################
# TASK: Extract relevant columns from mosaic fp adults data for regression
# clinical and sig meds
########################################################################

########################################################################
clinical_cols_data = c("mosaic"
                    , "ia_exac_copd"
                    , "death"
                    #, "obese2" #inc peaeds, but once you subset data for adults, its the same!
                    , "obesity"
                    , "flustat"
                    , "sfluv"
                    , "h1n1v"
                    , "age"
                   , "gender"
                    , "asthma"
                    , "vl_pfu_ul_npa1"
                    , "los"
                    , "onset2final"
                    , "onsfindeath"
                    , "onset_2_initial"
                    , "o2_sat_admis"
                    #, "o2_sat_suppl"
                    , "ethnicity"
                    , "smoking"
                    , "ia_cxr"
                    , "max_resp_score"
                    , "T1_resp_score"
                    , "com_noasthma"
                    , "T2_resp_score"
                    , "inresp_sev"
                    , "steroid")

clinical_cols_added = c("age_bins"
                          , "o2_sat_bin"
                          , "onset_initial_bin"
                          , "steroid_ics"
                          , "t1_resp_recoded" )

clinical_cols = c(clinical_cols_data, clinical_cols_added)
  
sig_npa_cols = c("eotaxin_npa1"
                 , "eotaxin3_npa1"
                 , "eotaxin3_npa2"
                 , "gmcsf_npa2"
                 , "ifnb_npa1"
                 , "ifnb_npa2"
                 , "il1_npa1"
                 , "il1_npa2"
                 , "il10_npa1"
                 , "il12p70_npa1"
                 , "il12p70_npa2"
                 , "il13_npa1"
                 , "il2_npa3"
                 , "il4_npa1"
                 , "il5_npa1"
                 , "il5_npa2"
                 , "il6_npa1"
                 , "il6_npa2"
                 , "il8_npa1"
                 , "il8_npa2"
                 , "il8_2_npa1"
                 , "il8_2_npa2"
                 , "ip10_npa1"
                 , "mcp4_npa1"
                 , "mcp4_npa2"
                 , "mdc_npa1"
                 , "mip17_npa1"
                 , "mip17_npa2"
                 , "neopterin_npa1"
                 , "rantes_npa1"
                 , "tarc_npa2"
                 , "tnf_npa2"
                 , "tnfr1_npa1" 
                 , "tnfr1_npa2"
                 , "tnfr1_npa3" 
                 , "tnfr2_npa1"
                 , "tnfr2_npa2"
                 , "tnfr2_npa3")

cols_to_extract = c(clinical_cols, sig_npa_cols)

if ( length(cols_to_extract) == length(clinical_cols) + length(sig_npa_cols) ){
  cat("PASS: length match"
      , "\nTotal no. of cols to extract for regression:", length(cols_to_extract)
      , "\nNo. of clinical cols:", length(clinical_cols)
      , "\nNo. of sig npa cols: ", length(sig_npa_cols))
}else{
  cat("FAIL: length mismatch"
      , "\nAborting!")
  quit()
}