mosaic_2020/colnames_clinical_meds.R

107 lines
3.6 KiB
R
Executable file

#!/usr/bin/Rscript
getwd()
setwd('~/git/mosaic_2020/')
getwd()
########################################################################
# TASK: Extract relevant columns from mosaic fp adults data for regression
# clinical and sig meds
########################################################################
########################################################################
# meta data columns
meta_data_cols = c("mosaic"
, "gender"
, "age"
, "adult"
, "flustat"
, "type"
#, "obese2" #inc peaeds, but once you subset data for adults, its the same!
#, "height", "height_unit"
#, "weight", "weight_unit"
#, "ia_height_ftin", "ia_height_m", "ia_weight"
#, "visual_est_bmi", "bmi_rating"
, "obesity")
clinical_cols = c("death"
, "sfluv"
, "h1n1v"
, "ia_exac_copd"
, "onset_2_t1"
, "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")
meta_clinical_cols = c(meta_data_cols, clinical_cols)
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, clinical_cols_added)
#if ( length(cols_to_extract) == length(clinical_cols) + length(sig_npa_cols) + length(clinical_cols_added) ){
# 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()
#}