DataSetup.Rmd

---
title: "FYP Redo"
author: "Lucy Whitmore"
date: "5/19/2022"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

packages <-  c("tidyverse",
               "reshape2",
               "nlme", "lme4",
               "data.table", "psych",
               "parallel","lubridate",
               "ggpubr", "broom", 
               "apaTables", "MetBrewer")
if (length(setdiff(packages, rownames(installed.packages()))) > 0) {
  install.packages(setdiff(packages, rownames(installed.packages())))  
}
lapply(packages, library, character.only = TRUE)
```

Load Structural Data
```{r}
#Load list of included subjects
struc_include_baseline<-rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_imgincl01.txt") %>%
  filter(!collection_title=="collection_title") %>% 
  select(1:11) %>% 
  mutate(interview_age = as.numeric(interview_age),
        imgincl_t1w_include = as.numeric(imgincl_t1w_include)) %>% 
  filter(imgincl_t1w_include == 1) %>% 
  filter(eventname=="baseline_year_1_arm_1")
  
  
struc_include_followup<-rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_imgincl01.txt") %>%
  filter(!collection_title=="collection_title") %>% 
  select(1:11) %>% 
  mutate(interview_age = as.numeric(interview_age),
        imgincl_t1w_include = as.numeric(imgincl_t1w_include)) %>% 
  filter(imgincl_t1w_include == 1) %>% 
  filter(eventname=="2_year_follow_up_y_arm_1")


#Load structural data
#Filter for volume & area measurements
smri_baseline <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_smrip10201.txt") %>%
  filter(eventname=="baseline_year_1_arm_1") %>% 
  filter(!collection_title=="collection_title") %>% 
  filter(src_subject_id%in%struc_include_baseline$src_subject_id) %>% 
  select(subjectkey, src_subject_id, interview_age, sex, eventname, matches("vol|area")) %>% 
  select(-contains(c('cf'))) %>% #remove genetically derived parcellations
  select(-c(smri_area_cdk_totallh, smri_area_cdk_totalrh, smri_area_cdk_total, smri_vol_scs_lesionlh, smri_vol_scs_lesionrh, smri_vol_scs_wmhint, smri_vol_scs_wmhintlh, smri_vol_scs_wmhintrh, smri_vol_scs_wholeb, smri_vol_scs_latventricles, smri_vol_scs_allventricles, smri_vol_scs_cbwmatterrh, smri_vol_scs_cbwmatterlh))



smri_followup <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_smrip10201.txt") %>%
  filter(eventname=="2_year_follow_up_y_arm_1") %>% 
  filter(!collection_title=="collection_title") %>% 
  filter(src_subject_id%in%struc_include_followup$src_subject_id) %>% 
  select(subjectkey, src_subject_id, interview_age, sex, eventname, matches("vol|area")) %>% 
  select(-contains(c('cf'))) %>% #remove genetically derived parcellations
  select(-c(smri_area_cdk_totallh, smri_area_cdk_totalrh, smri_area_cdk_total, smri_vol_scs_lesionlh, smri_vol_scs_lesionrh, smri_vol_scs_wmhint, smri_vol_scs_wmhintlh, smri_vol_scs_wmhintrh, smri_vol_scs_wholeb, smri_vol_scs_latventricles, smri_vol_scs_allventricles, smri_vol_scs_cbwmatterrh, smri_vol_scs_cbwmatterlh))


#renamed_smri_baseline<-rio::import("/Volumes/devbrainlab/Lucy/BrainAGE/FYP/renamed_smri_baseline.Rda")

```


Rename Columns Baseline
```{r}
renamed_smri_baseline <- smri_baseline %>% 
  rename(FS_InterCranial_Vol = smri_vol_scs_intracranialv, #technically inter vs intra
#FS_BrainSeg_Vol
#FS_BrainSeg_Vol_No_Vent
#FS_BrainSeg_Vol_No_Vent_Surf
FS_LCort_GM_Vol = smri_vol_cdk_totallh,
FS_RCort_GM_Vol = smri_vol_cdk_totalrh,
FS_TotCort_GM_Vol = smri_vol_cdk_total, #total whole brain cortical volume
FS_SubCort_GM_Vol = smri_vol_scs_subcorticalgv,
#FS_Total_GM_Vol #maybe just calculate?
FS_SupraTentorial_Vol = smri_vol_scs_suprateialv, #assuming this includes all ventricles
#FS_SupraTentorial_Vol_No_Vent  #smri_vol_scs_allventricles is all ventricles, subtract?
#FS_SupraTentorial_No_Vent_Voxel_Count
#FS_Mask_Vol
#FS_BrainSegVol_eTIV_Ratio
#FS_MaskVol_eTIV_Ratio
FS_L_LatVent_Vol = smri_vol_scs_ltventriclelh,
FS_L_InfLatVent_Vol = smri_vol_scs_inflatventlh,
FS_L_Cerebellum_WM_Vol = smri_vol_scs_crbwmatterlh,
FS_L_Cerebellum_Cort_Vol = smri_vol_scs_crbcortexlh,
FS_L_ThalamusProper_Vol = smri_vol_scs_tplh,
FS_L_Caudate_Vol = smri_vol_scs_caudatelh,
FS_L_Putamen_Vol = smri_vol_scs_putamenlh,
FS_L_Pallidum_Vol = smri_vol_scs_pallidumlh,
FS_3rdVent_Vol = smri_vol_scs_3rdventricle,
FS_4thVent_Vol = smri_vol_scs_4thventricle,
FS_BrainStem_Vol = smri_vol_scs_bstem,
FS_L_Hippo_Vol = smri_vol_scs_hpuslh,
FS_L_Amygdala_Vol =	smri_vol_scs_amygdalalh,
FS_CSF_Vol = smri_vol_scs_csf,
FS_L_AccumbensArea_Vol = smri_vol_scs_aal,
FS_L_VentDC_Vol = smri_vol_scs_vedclh,
#FS_L_Vessel_Vol
#FS_L_ChoroidPlexus_Vol
FS_R_LatVent_Vol = 	smri_vol_scs_ltventriclerh,
FS_R_InfLatVent_Vol = smri_vol_scs_inflatventrh,
FS_R_Cerebellum_WM_Vol = smri_vol_scs_crbwmatterrh,
FS_R_Cerebellum_Cort_Vol = smri_vol_scs_crbcortexrh,
FS_R_ThalamusProper_Vol = smri_vol_scs_tprh,
FS_R_Caudate_Vol = smri_vol_scs_caudaterh,
FS_R_Putamen_Vol = smri_vol_scs_putamenrh,
FS_R_Pallidum_Vol = smri_vol_scs_pallidumrh,
FS_R_Hippo_Vol = smri_vol_scs_hpusrh,
FS_R_Amygdala_Vol = smri_vol_scs_amygdalarh,
FS_R_AccumbensArea_Vol = 	smri_vol_scs_aar,
FS_R_VentDC_Vol = smri_vol_scs_vedcrh,
#FS_R_Vessel_Vol
#FS_R_ChoroidPlexus_Vol
#FS_OpticChiasm_Vol
FS_CC_Posterior_Vol = smri_vol_scs_ccps,
FS_CC_MidPosterior_Vol = smri_vol_scs_ccmidps,
FS_CC_Central_Vol = smri_vol_scs_ccct,
FS_CC_MidAnterior_Vol = smri_vol_scs_ccmidat,
FS_CC_Anterior_Vol = smri_vol_scs_ccat,
FS_L_Bankssts_Area = 	smri_area_cdk_banksstslh,
FS_L_Caudalanteriorcingulate_Area = smri_area_cdk_cdacatelh,
FS_L_Caudalmiddlefrontal_Area = smri_area_cdk_cdmdfrlh,
FS_L_Cuneus_Area = 	smri_area_cdk_cuneuslh,
FS_L_Entorhinal_Area = smri_area_cdk_ehinallh,
FS_L_Fusiform_Area = smri_area_cdk_fusiformlh,
FS_L_Inferiorparietal_Area = smri_area_cdk_ifpllh,
FS_L_Inferiortemporal_Area = smri_area_cdk_iftmlh,
FS_L_Isthmuscingulate_Area = smri_area_cdk_ihcatelh,
FS_L_Lateraloccipital_Area = smri_area_cdk_locclh,
FS_L_Lateralorbitofrontal_Area = smri_area_cdk_lobfrlh,
FS_L_Lingual_Area = smri_area_cdk_linguallh,
FS_L_Medialorbitofrontal_Area = smri_area_cdk_mobfrlh,
FS_L_Middletemporal_Area = smri_area_cdk_mdtmlh,
FS_L_Parahippocampal_Area = smri_area_cdk_parahpallh,
FS_L_Paracentral_Area = smri_area_cdk_paracnlh,
FS_L_Parsopercularis_Area = smri_area_cdk_parsopclh,
FS_L_Parsorbitalis_Area = smri_area_cdk_parsobislh,
FS_L_Parstriangularis_Area = smri_area_cdk_parstgrislh,
FS_L_Pericalcarine_Area = smri_area_cdk_pericclh,
FS_L_Postcentral_Area = smri_area_cdk_postcnlh,
FS_L_Posteriorcingulate_Area = smri_area_cdk_ptcatelh,
FS_L_Precentral_Area = smri_area_cdk_precnlh,
FS_L_Precuneus_Area = smri_area_cdk_pclh,
FS_L_Rostralanteriorcingulate_Area = smri_area_cdk_rracatelh,
FS_L_Rostralmiddlefrontal_Area = smri_area_cdk_rrmdfrlh,
FS_L_Superiorfrontal_Area = smri_area_cdk_sufrlh,
FS_L_Superiorparietal_Area = smri_area_cdk_supllh,
FS_L_Superiortemporal_Area = smri_area_cdk_sutmlh,
FS_L_Supramarginal_Area = smri_area_cdk_smlh,
FS_L_Frontalpole_Area = smri_area_cdk_frpolelh,
FS_L_Temporalpole_Area = smri_area_cdk_tmpolelh,
FS_L_Transversetemporal_Area = smri_area_cdk_trvtmlh,
FS_L_Insula_Area = smri_area_cdk_insulalh,
FS_R_Bankssts_Area = smri_area_cdk_banksstsrh,
FS_R_Caudalanteriorcingulate_Area = smri_area_cdk_cdacaterh,
FS_R_Caudalmiddlefrontal_Area = smri_area_cdk_cdmdfrrh,
FS_R_Cuneus_Area = smri_area_cdk_cuneusrh,
FS_R_Entorhinal_Area = smri_area_cdk_ehinalrh,
FS_R_Fusiform_Area = smri_area_cdk_fusiformrh,
FS_R_Inferiorparietal_Area = smri_area_cdk_ifplrh,
FS_R_Inferiortemporal_Area = 	smri_area_cdk_iftmrh,
FS_R_Isthmuscingulate_Area = smri_area_cdk_ihcaterh,
FS_R_Lateraloccipital_Area = smri_area_cdk_loccrh,
FS_R_Lateralorbitofrontal_Area = smri_area_cdk_lobfrrh,
FS_R_Lingual_Area = smri_area_cdk_lingualrh,
FS_R_Medialorbitofrontal_Area = smri_area_cdk_mobfrrh,
FS_R_Middletemporal_Area = smri_area_cdk_mdtmrh,
FS_R_Parahippocampal_Area = smri_area_cdk_parahpalrh,
FS_R_Paracentral_Area = smri_area_cdk_paracnrh,
FS_R_Parsopercularis_Area = smri_area_cdk_parsopcrh,
FS_R_Parsorbitalis_Area = smri_area_cdk_parsobisrh,
FS_R_Parstriangularis_Area = smri_area_cdk_parstgrisrh,
FS_R_Pericalcarine_Area = smri_area_cdk_periccrh,
FS_R_Postcentral_Area = smri_area_cdk_postcnrh,
FS_R_Posteriorcingulate_Area = smri_area_cdk_ptcaterh,
FS_R_Precentral_Area = smri_area_cdk_precnrh,
FS_R_Precuneus_Area = smri_area_cdk_pcrh,
FS_R_Rostralanteriorcingulate_Area = smri_area_cdk_rracaterh,
FS_R_Rostralmiddlefrontal_Area = smri_area_cdk_rrmdfrrh,
FS_R_Superiorfrontal_Area = smri_area_cdk_sufrrh,
FS_R_Superiorparietal_Area = 	smri_area_cdk_suplrh,
FS_R_Superiortemporal_Area = smri_area_cdk_sutmrh,
FS_R_Supramarginal_Area = smri_area_cdk_smrh,
FS_R_Frontalpole_Area = smri_area_cdk_frpolerh,
FS_R_Temporalpole_Area = smri_area_cdk_tmpolerh,
FS_R_Transversetemporal_Area = smri_area_cdk_trvtmrh,
FS_R_Insula_Area = smri_area_cdk_insularh,
#abcd technically doesn't say gray matter volume for the following varibles
FS_L_Bankssts_GrayVol = smri_vol_cdk_banksstslh,
FS_L_Caudalanteriorcingulate_GrayVol = smri_vol_cdk_cdacatelh,
FS_L_Caudalmiddlefrontal_GrayVol = smri_vol_cdk_cdmdfrlh,
FS_L_Cuneus_GrayVol = smri_vol_cdk_cuneuslh,
FS_L_Entorhinal_GrayVol = smri_vol_cdk_ehinallh,
FS_L_Fusiform_GrayVol = smri_vol_cdk_fusiformlh,
FS_L_Inferiorparietal_GrayVol = smri_vol_cdk_ifpllh,
FS_L_Inferiortemporal_GrayVol = smri_vol_cdk_iftmlh,
FS_L_Isthmuscingulate_GrayVol = smri_vol_cdk_ihcatelh,
FS_L_Lateraloccipital_GrayVol = smri_vol_cdk_locclh,
FS_L_Lateralorbitofrontal_GrayVol = smri_vol_cdk_lobfrlh,
FS_L_Lingual_GrayVol = smri_vol_cdk_linguallh,
FS_L_Medialorbitofrontal_GrayVol = smri_vol_cdk_mobfrlh,
FS_L_Middletemporal_GrayVol = smri_vol_cdk_mdtmlh,
FS_L_Parahippocampal_GrayVol = smri_vol_cdk_parahpallh,
FS_L_Paracentral_GrayVol = smri_vol_cdk_paracnlh,
FS_L_Parsopercularis_GrayVol = smri_vol_cdk_parsopclh,
FS_L_Parsorbitalis_GrayVol = smri_vol_cdk_parsobislh,
FS_L_Parstriangularis_GrayVol = smri_vol_cdk_parstgrislh,
FS_L_Pericalcarine_GrayVol = smri_vol_cdk_pericclh,
FS_L_Postcentral_GrayVol = smri_vol_cdk_postcnlh,
FS_L_Posteriorcingulate_GrayVol = smri_vol_cdk_ptcatelh,
FS_L_Precentral_GrayVol = smri_vol_cdk_precnlh,
FS_L_Precuneus_GrayVol = smri_vol_cdk_pclh,
FS_L_Rostralanteriorcingulate_GrayVol = smri_vol_cdk_rracatelh,
FS_L_Rostralmiddlefrontal_GrayVol = smri_vol_cdk_rrmdfrlh,
FS_L_Superiorfrontal_GrayVol = smri_vol_cdk_sufrlh,
FS_L_Superiorparietal_GrayVol = smri_vol_cdk_supllh,
FS_L_Superiortemporal_GrayVol = smri_vol_cdk_sutmlh,
FS_L_Supramarginal_GrayVol = smri_vol_cdk_smlh,
FS_L_Frontalpole_GrayVol = smri_vol_cdk_frpolelh,
FS_L_Temporalpole_GrayVol = smri_vol_cdk_tmpolelh,
FS_L_Transversetemporal_GrayVol = smri_vol_cdk_trvtmlh,
FS_L_Insula_GrayVol = smri_vol_cdk_insulalh,
FS_R_Bankssts_GrayVol = smri_vol_cdk_banksstsrh,
FS_R_Caudalanteriorcingulate_GrayVol = smri_vol_cdk_cdacaterh,
FS_R_Caudalmiddlefrontal_GrayVol = 	smri_vol_cdk_cdmdfrrh,
FS_R_Cuneus_GrayVol = smri_vol_cdk_cuneusrh,
FS_R_Entorhinal_GrayVol = smri_vol_cdk_ehinalrh,
FS_R_Fusiform_GrayVol = smri_vol_cdk_fusiformrh,
FS_R_Inferiorparietal_GrayVol = smri_vol_cdk_ifplrh,
FS_R_Inferiortemporal_GrayVol = smri_vol_cdk_iftmrh,
FS_R_Isthmuscingulate_GrayVol = smri_vol_cdk_ihcaterh,
FS_R_Lateraloccipital_GrayVol = smri_vol_cdk_loccrh,
FS_R_Lateralorbitofrontal_GrayVol = smri_vol_cdk_lobfrrh,
FS_R_Lingual_GrayVol = smri_vol_cdk_lingualrh,
FS_R_Medialorbitofrontal_GrayVol = smri_vol_cdk_mobfrrh,
FS_R_Middletemporal_GrayVol = smri_vol_cdk_mdtmrh,
FS_R_Parahippocampal_GrayVol = smri_vol_cdk_parahpalrh,
FS_R_Paracentral_GrayVol = smri_vol_cdk_paracnrh,
FS_R_Parsopercularis_GrayVol = smri_vol_cdk_parsopcrh,
FS_R_Parsorbitalis_GrayVol = smri_vol_cdk_parsobisrh,
FS_R_Parstriangularis_GrayVol = smri_vol_cdk_parstgrisrh,
FS_R_Pericalcarine_GrayVol = smri_vol_cdk_periccrh,
FS_R_Postcentral_GrayVol = smri_vol_cdk_postcnrh,
FS_R_Posteriorcingulate_GrayVol = 	smri_vol_cdk_ptcaterh,
FS_R_Precentral_GrayVol = smri_vol_cdk_precnrh,
FS_R_Precuneus_GrayVol = smri_vol_cdk_pcrh,
FS_R_Rostralanteriorcingulate_GrayVol = smri_vol_cdk_rracaterh,
FS_R_Rostralmiddlefrontal_GrayVol = smri_vol_cdk_rrmdfrrh,
FS_R_Superiorfrontal_GrayVol = smri_vol_cdk_sufrrh,
FS_R_Superiorparietal_GrayVol = smri_vol_cdk_suplrh,
FS_R_Superiortemporal_GrayVol = smri_vol_cdk_sutmrh,
FS_R_Supramarginal_GrayVol = smri_vol_cdk_smrh,
FS_R_Frontalpole_GrayVol = smri_vol_cdk_frpolerh,
FS_R_Temporalpole_GrayVol = smri_vol_cdk_tmpolerh,
FS_R_Transversetemporal_GrayVol = smri_vol_cdk_trvtmrh,
FS_R_Insula_GrayVol = smri_vol_cdk_insularh
)


#Create empty columns to match feature list
renamed_smri_baseline[,"FS_BrainSeg_Vol"] <- NA
renamed_smri_baseline[,"FS_BrainSeg_Vol_No_Vent"] <- NA
renamed_smri_baseline[,"FS_BrainSeg_Vol_No_Vent_Surf"] <- NA
renamed_smri_baseline[,"FS_SupraTentorial_Vol_No_Vent"] <- NA
renamed_smri_baseline[,"FS_SupraTentorial_No_Vent_Voxel_Count"] <- NA
renamed_smri_baseline[,"FS_Mask_Vol"] <- NA
renamed_smri_baseline[,"FS_BrainSegVol_eTIV_Ratio"] <- NA
renamed_smri_baseline[,"FS_MaskVol_eTIV_Ratio"] <- NA
renamed_smri_baseline[,"FS_L_Vessel_Vol"] <- NA
renamed_smri_baseline[,"FS_L_ChoroidPlexus_Vol"] <- NA
renamed_smri_baseline[,"FS_R_Vessel_Vol"] <- NA
renamed_smri_baseline[,"FS_R_ChoroidPlexus_Vol"] <- NA
renamed_smri_baseline[,"FS_OpticChiasm_Vol"] <- NA
renamed_smri_baseline[,"FS_Total_GM_Vol"] <- NA


#Convert to numeric
renamed_smri_baseline[,6:ncol(renamed_smri_baseline)] <- sapply(renamed_smri_baseline[,6:ncol(renamed_smri_baseline)],as.numeric)
renamed_smri_baseline <- renamed_smri_baseline %>% 
  mutate(interview_age = as.numeric(interview_age)/12)

save(renamed_smri_baseline, file="renamed_smri_baseline.Rda")


smri_abcd_model_baseline <- renamed_smri_baseline[, 1:180] 
```


Rename Columns Follow-up
```{r}
renamed_smri_followup <- smri_followup %>% 
  rename(FS_InterCranial_Vol = smri_vol_scs_intracranialv, #technically inter vs intra
#FS_BrainSeg_Vol
#FS_BrainSeg_Vol_No_Vent
#FS_BrainSeg_Vol_No_Vent_Surf
FS_LCort_GM_Vol = smri_vol_cdk_totallh,
FS_RCort_GM_Vol = smri_vol_cdk_totalrh,
FS_TotCort_GM_Vol = smri_vol_cdk_total, #total whole brain cortical volume
FS_SubCort_GM_Vol = smri_vol_scs_subcorticalgv,
#FS_Total_GM_Vol #maybe just calculate?
FS_SupraTentorial_Vol = smri_vol_scs_suprateialv, #assuming this includes all ventricles
#FS_SupraTentorial_Vol_No_Vent  #smri_vol_scs_allventricles is all ventricles, subtract?
#FS_SupraTentorial_No_Vent_Voxel_Count
#FS_Mask_Vol
#FS_BrainSegVol_eTIV_Ratio
#FS_MaskVol_eTIV_Ratio
FS_L_LatVent_Vol = smri_vol_scs_ltventriclelh,
FS_L_InfLatVent_Vol = smri_vol_scs_inflatventlh,
FS_L_Cerebellum_WM_Vol = smri_vol_scs_crbwmatterlh,
FS_L_Cerebellum_Cort_Vol = smri_vol_scs_crbcortexlh,
FS_L_ThalamusProper_Vol = smri_vol_scs_tplh,
FS_L_Caudate_Vol = smri_vol_scs_caudatelh,
FS_L_Putamen_Vol = smri_vol_scs_putamenlh,
FS_L_Pallidum_Vol = smri_vol_scs_pallidumlh,
FS_3rdVent_Vol = smri_vol_scs_3rdventricle,
FS_4thVent_Vol = smri_vol_scs_4thventricle,
FS_BrainStem_Vol = smri_vol_scs_bstem,
FS_L_Hippo_Vol = smri_vol_scs_hpuslh,
FS_L_Amygdala_Vol =	smri_vol_scs_amygdalalh,
FS_CSF_Vol = smri_vol_scs_csf,
FS_L_AccumbensArea_Vol = smri_vol_scs_aal,
FS_L_VentDC_Vol = smri_vol_scs_vedclh,
#FS_L_Vessel_Vol
#FS_L_ChoroidPlexus_Vol
FS_R_LatVent_Vol = 	smri_vol_scs_ltventriclerh,
FS_R_InfLatVent_Vol = smri_vol_scs_inflatventrh,
FS_R_Cerebellum_WM_Vol = smri_vol_scs_crbwmatterrh,
FS_R_Cerebellum_Cort_Vol = smri_vol_scs_crbcortexrh,
FS_R_ThalamusProper_Vol = smri_vol_scs_tprh,
FS_R_Caudate_Vol = smri_vol_scs_caudaterh,
FS_R_Putamen_Vol = smri_vol_scs_putamenrh,
FS_R_Pallidum_Vol = smri_vol_scs_pallidumrh,
FS_R_Hippo_Vol = smri_vol_scs_hpusrh,
FS_R_Amygdala_Vol = smri_vol_scs_amygdalarh,
FS_R_AccumbensArea_Vol = 	smri_vol_scs_aar,
FS_R_VentDC_Vol = smri_vol_scs_vedcrh,
#FS_R_Vessel_Vol
#FS_R_ChoroidPlexus_Vol
#FS_OpticChiasm_Vol
FS_CC_Posterior_Vol = smri_vol_scs_ccps,
FS_CC_MidPosterior_Vol = smri_vol_scs_ccmidps,
FS_CC_Central_Vol = smri_vol_scs_ccct,
FS_CC_MidAnterior_Vol = smri_vol_scs_ccmidat,
FS_CC_Anterior_Vol = smri_vol_scs_ccat,
FS_L_Bankssts_Area = 	smri_area_cdk_banksstslh,
FS_L_Caudalanteriorcingulate_Area = smri_area_cdk_cdacatelh,
FS_L_Caudalmiddlefrontal_Area = smri_area_cdk_cdmdfrlh,
FS_L_Cuneus_Area = 	smri_area_cdk_cuneuslh,
FS_L_Entorhinal_Area = smri_area_cdk_ehinallh,
FS_L_Fusiform_Area = smri_area_cdk_fusiformlh,
FS_L_Inferiorparietal_Area = smri_area_cdk_ifpllh,
FS_L_Inferiortemporal_Area = smri_area_cdk_iftmlh,
FS_L_Isthmuscingulate_Area = smri_area_cdk_ihcatelh,
FS_L_Lateraloccipital_Area = smri_area_cdk_locclh,
FS_L_Lateralorbitofrontal_Area = smri_area_cdk_lobfrlh,
FS_L_Lingual_Area = smri_area_cdk_linguallh,
FS_L_Medialorbitofrontal_Area = smri_area_cdk_mobfrlh,
FS_L_Middletemporal_Area = smri_area_cdk_mdtmlh,
FS_L_Parahippocampal_Area = smri_area_cdk_parahpallh,
FS_L_Paracentral_Area = smri_area_cdk_paracnlh,
FS_L_Parsopercularis_Area = smri_area_cdk_parsopclh,
FS_L_Parsorbitalis_Area = smri_area_cdk_parsobislh,
FS_L_Parstriangularis_Area = smri_area_cdk_parstgrislh,
FS_L_Pericalcarine_Area = smri_area_cdk_pericclh,
FS_L_Postcentral_Area = smri_area_cdk_postcnlh,
FS_L_Posteriorcingulate_Area = smri_area_cdk_ptcatelh,
FS_L_Precentral_Area = smri_area_cdk_precnlh,
FS_L_Precuneus_Area = smri_area_cdk_pclh,
FS_L_Rostralanteriorcingulate_Area = smri_area_cdk_rracatelh,
FS_L_Rostralmiddlefrontal_Area = smri_area_cdk_rrmdfrlh,
FS_L_Superiorfrontal_Area = smri_area_cdk_sufrlh,
FS_L_Superiorparietal_Area = smri_area_cdk_supllh,
FS_L_Superiortemporal_Area = smri_area_cdk_sutmlh,
FS_L_Supramarginal_Area = smri_area_cdk_smlh,
FS_L_Frontalpole_Area = smri_area_cdk_frpolelh,
FS_L_Temporalpole_Area = smri_area_cdk_tmpolelh,
FS_L_Transversetemporal_Area = smri_area_cdk_trvtmlh,
FS_L_Insula_Area = smri_area_cdk_insulalh,
FS_R_Bankssts_Area = smri_area_cdk_banksstsrh,
FS_R_Caudalanteriorcingulate_Area = smri_area_cdk_cdacaterh,
FS_R_Caudalmiddlefrontal_Area = smri_area_cdk_cdmdfrrh,
FS_R_Cuneus_Area = smri_area_cdk_cuneusrh,
FS_R_Entorhinal_Area = smri_area_cdk_ehinalrh,
FS_R_Fusiform_Area = smri_area_cdk_fusiformrh,
FS_R_Inferiorparietal_Area = smri_area_cdk_ifplrh,
FS_R_Inferiortemporal_Area = 	smri_area_cdk_iftmrh,
FS_R_Isthmuscingulate_Area = smri_area_cdk_ihcaterh,
FS_R_Lateraloccipital_Area = smri_area_cdk_loccrh,
FS_R_Lateralorbitofrontal_Area = smri_area_cdk_lobfrrh,
FS_R_Lingual_Area = smri_area_cdk_lingualrh,
FS_R_Medialorbitofrontal_Area = smri_area_cdk_mobfrrh,
FS_R_Middletemporal_Area = smri_area_cdk_mdtmrh,
FS_R_Parahippocampal_Area = smri_area_cdk_parahpalrh,
FS_R_Paracentral_Area = smri_area_cdk_paracnrh,
FS_R_Parsopercularis_Area = smri_area_cdk_parsopcrh,
FS_R_Parsorbitalis_Area = smri_area_cdk_parsobisrh,
FS_R_Parstriangularis_Area = smri_area_cdk_parstgrisrh,
FS_R_Pericalcarine_Area = smri_area_cdk_periccrh,
FS_R_Postcentral_Area = smri_area_cdk_postcnrh,
FS_R_Posteriorcingulate_Area = smri_area_cdk_ptcaterh,
FS_R_Precentral_Area = smri_area_cdk_precnrh,
FS_R_Precuneus_Area = smri_area_cdk_pcrh,
FS_R_Rostralanteriorcingulate_Area = smri_area_cdk_rracaterh,
FS_R_Rostralmiddlefrontal_Area = smri_area_cdk_rrmdfrrh,
FS_R_Superiorfrontal_Area = smri_area_cdk_sufrrh,
FS_R_Superiorparietal_Area = 	smri_area_cdk_suplrh,
FS_R_Superiortemporal_Area = smri_area_cdk_sutmrh,
FS_R_Supramarginal_Area = smri_area_cdk_smrh,
FS_R_Frontalpole_Area = smri_area_cdk_frpolerh,
FS_R_Temporalpole_Area = smri_area_cdk_tmpolerh,
FS_R_Transversetemporal_Area = smri_area_cdk_trvtmrh,
FS_R_Insula_Area = smri_area_cdk_insularh,
#abcd technically doesn't say gray matter volume for the following varibles
FS_L_Bankssts_GrayVol = smri_vol_cdk_banksstslh,
FS_L_Caudalanteriorcingulate_GrayVol = smri_vol_cdk_cdacatelh,
FS_L_Caudalmiddlefrontal_GrayVol = smri_vol_cdk_cdmdfrlh,
FS_L_Cuneus_GrayVol = smri_vol_cdk_cuneuslh,
FS_L_Entorhinal_GrayVol = smri_vol_cdk_ehinallh,
FS_L_Fusiform_GrayVol = smri_vol_cdk_fusiformlh,
FS_L_Inferiorparietal_GrayVol = smri_vol_cdk_ifpllh,
FS_L_Inferiortemporal_GrayVol = smri_vol_cdk_iftmlh,
FS_L_Isthmuscingulate_GrayVol = smri_vol_cdk_ihcatelh,
FS_L_Lateraloccipital_GrayVol = smri_vol_cdk_locclh,
FS_L_Lateralorbitofrontal_GrayVol = smri_vol_cdk_lobfrlh,
FS_L_Lingual_GrayVol = smri_vol_cdk_linguallh,
FS_L_Medialorbitofrontal_GrayVol = smri_vol_cdk_mobfrlh,
FS_L_Middletemporal_GrayVol = smri_vol_cdk_mdtmlh,
FS_L_Parahippocampal_GrayVol = smri_vol_cdk_parahpallh,
FS_L_Paracentral_GrayVol = smri_vol_cdk_paracnlh,
FS_L_Parsopercularis_GrayVol = smri_vol_cdk_parsopclh,
FS_L_Parsorbitalis_GrayVol = smri_vol_cdk_parsobislh,
FS_L_Parstriangularis_GrayVol = smri_vol_cdk_parstgrislh,
FS_L_Pericalcarine_GrayVol = smri_vol_cdk_pericclh,
FS_L_Postcentral_GrayVol = smri_vol_cdk_postcnlh,
FS_L_Posteriorcingulate_GrayVol = smri_vol_cdk_ptcatelh,
FS_L_Precentral_GrayVol = smri_vol_cdk_precnlh,
FS_L_Precuneus_GrayVol = smri_vol_cdk_pclh,
FS_L_Rostralanteriorcingulate_GrayVol = smri_vol_cdk_rracatelh,
FS_L_Rostralmiddlefrontal_GrayVol = smri_vol_cdk_rrmdfrlh,
FS_L_Superiorfrontal_GrayVol = smri_vol_cdk_sufrlh,
FS_L_Superiorparietal_GrayVol = smri_vol_cdk_supllh,
FS_L_Superiortemporal_GrayVol = smri_vol_cdk_sutmlh,
FS_L_Supramarginal_GrayVol = smri_vol_cdk_smlh,
FS_L_Frontalpole_GrayVol = smri_vol_cdk_frpolelh,
FS_L_Temporalpole_GrayVol = smri_vol_cdk_tmpolelh,
FS_L_Transversetemporal_GrayVol = smri_vol_cdk_trvtmlh,
FS_L_Insula_GrayVol = smri_vol_cdk_insulalh,
FS_R_Bankssts_GrayVol = smri_vol_cdk_banksstsrh,
FS_R_Caudalanteriorcingulate_GrayVol = smri_vol_cdk_cdacaterh,
FS_R_Caudalmiddlefrontal_GrayVol = 	smri_vol_cdk_cdmdfrrh,
FS_R_Cuneus_GrayVol = smri_vol_cdk_cuneusrh,
FS_R_Entorhinal_GrayVol = smri_vol_cdk_ehinalrh,
FS_R_Fusiform_GrayVol = smri_vol_cdk_fusiformrh,
FS_R_Inferiorparietal_GrayVol = smri_vol_cdk_ifplrh,
FS_R_Inferiortemporal_GrayVol = smri_vol_cdk_iftmrh,
FS_R_Isthmuscingulate_GrayVol = smri_vol_cdk_ihcaterh,
FS_R_Lateraloccipital_GrayVol = smri_vol_cdk_loccrh,
FS_R_Lateralorbitofrontal_GrayVol = smri_vol_cdk_lobfrrh,
FS_R_Lingual_GrayVol = smri_vol_cdk_lingualrh,
FS_R_Medialorbitofrontal_GrayVol = smri_vol_cdk_mobfrrh,
FS_R_Middletemporal_GrayVol = smri_vol_cdk_mdtmrh,
FS_R_Parahippocampal_GrayVol = smri_vol_cdk_parahpalrh,
FS_R_Paracentral_GrayVol = smri_vol_cdk_paracnrh,
FS_R_Parsopercularis_GrayVol = smri_vol_cdk_parsopcrh,
FS_R_Parsorbitalis_GrayVol = smri_vol_cdk_parsobisrh,
FS_R_Parstriangularis_GrayVol = smri_vol_cdk_parstgrisrh,
FS_R_Pericalcarine_GrayVol = smri_vol_cdk_periccrh,
FS_R_Postcentral_GrayVol = smri_vol_cdk_postcnrh,
FS_R_Posteriorcingulate_GrayVol = 	smri_vol_cdk_ptcaterh,
FS_R_Precentral_GrayVol = smri_vol_cdk_precnrh,
FS_R_Precuneus_GrayVol = smri_vol_cdk_pcrh,
FS_R_Rostralanteriorcingulate_GrayVol = smri_vol_cdk_rracaterh,
FS_R_Rostralmiddlefrontal_GrayVol = smri_vol_cdk_rrmdfrrh,
FS_R_Superiorfrontal_GrayVol = smri_vol_cdk_sufrrh,
FS_R_Superiorparietal_GrayVol = smri_vol_cdk_suplrh,
FS_R_Superiortemporal_GrayVol = smri_vol_cdk_sutmrh,
FS_R_Supramarginal_GrayVol = smri_vol_cdk_smrh,
FS_R_Frontalpole_GrayVol = smri_vol_cdk_frpolerh,
FS_R_Temporalpole_GrayVol = smri_vol_cdk_tmpolerh,
FS_R_Transversetemporal_GrayVol = smri_vol_cdk_trvtmrh,
FS_R_Insula_GrayVol = smri_vol_cdk_insularh
)


#Create empty columns to match feature list
renamed_smri_followup[,"FS_BrainSeg_Vol"] <- NA
renamed_smri_followup[,"FS_BrainSeg_Vol_No_Vent"] <- NA
renamed_smri_followup[,"FS_BrainSeg_Vol_No_Vent_Surf"] <- NA
renamed_smri_followup[,"FS_SupraTentorial_Vol_No_Vent"] <- NA
renamed_smri_followup[,"FS_SupraTentorial_No_Vent_Voxel_Count"] <- NA
renamed_smri_followup[,"FS_Mask_Vol"] <- NA
renamed_smri_followup[,"FS_BrainSegVol_eTIV_Ratio"] <- NA
renamed_smri_followup[,"FS_MaskVol_eTIV_Ratio"] <- NA
renamed_smri_followup[,"FS_L_Vessel_Vol"] <- NA
renamed_smri_followup[,"FS_L_ChoroidPlexus_Vol"] <- NA
renamed_smri_followup[,"FS_R_Vessel_Vol"] <- NA
renamed_smri_followup[,"FS_R_ChoroidPlexus_Vol"] <- NA
renamed_smri_followup[,"FS_OpticChiasm_Vol"] <- NA
renamed_smri_followup[,"FS_Total_GM_Vol"] <- NA


#Convert to numeric
renamed_smri_followup[,6:ncol(renamed_smri_followup)] <- sapply(renamed_smri_followup[,6:ncol(renamed_smri_followup)],as.numeric)
renamed_smri_followup <- renamed_smri_followup %>% 
  mutate(interview_age = as.numeric(interview_age)/12)

save(renamed_smri_followup, file="renamed_smri_followup.Rda")


smri_abcd_model_followup <- renamed_smri_followup[, 1:180] 
```


Load Other Variables
```{r}
#Puberty
#Raw scores, youth
youth_pds <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_ypdms01.txt") %>%
  filter(!collection_title=="collection_title") %>% 
  select(5:38) %>% 
  select(-c(interview_date, pds_remote___1, pds_remote___2, pds_remote___3, pds_remote___4, pds_device)) %>% 
  mutate_at(c(5:28), as.numeric)

#Raw scores, parent
parent_pds <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_ppdms01.txt") %>%
  filter(!collection_title=="collection_title")  %>% 
  select(5:27) %>% 
  select(-c(interview_date, pds_select_language___1)) %>% 
  mutate_at(c(5:21), as.numeric)

#Cognition Scores (Age-Corrected)
cognition <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_tbss01.txt") %>%
  filter(!collection_title=="collection_title")  %>% 
  select(src_subject_id,interview_age, eventname, nihtbx_fluidcomp_agecorrected, nihtbx_cryst_agecorrected, nihtbx_totalcomp_agecorrected, nihtbx_totalcomp_uncorrected) %>% 
  mutate_at(c(2, 4:7), as.numeric)

#nihtbx_totalcomp_rawscore total cognition raw score
#nihtbx_totalcomp_cs total cognition computed score
#	nihtbx_totalcomp_itmcnt cognition total composite item count
# might need to adjust for tasks completed?

#gish <- rio::import("/Volumes/devbrainlab/ABCD_Data/ABCD4pt0/abcd_gish2y01.txt") %>%
#  filter(!collection_title=="collection_title")  %>% 
 # filter(eventname == "2_year_follow_up_y_arm_1") %>% 
#   select(src_subject_id, sex, contains("gish2")) %>% 
 # select(-c(gish2_device, abcd_gish2y01_id))%>% 
 # mutate_at(c(3:15), as.numeric) 

#calculate gish score
```


## Format Puberty Data 
```{r}
#two or more missing values = exclude
#PDS mean score is a continuous value calculated by averaging across PDS items following Herting et al. (2021). Items on the PDS were considered missing if participants answered, “I don’t know” or “refuse to answer”, if the response was left blank, or if the response value was outside of the expected range (1 to 4). Participants with two or more missing answers were excluded from the PDS score calculation (following recommendations from (Herting et al., 2021)


### Youth-report, removes everyone w/o complete data
youth_pds[youth_pds == 777] <- NA
youth_pds[youth_pds == 999] <- NA


# Multiwave
youth_pds_multiwave <- youth_pds %>% 
  filter(eventname == "baseline_year_1_arm_1" | eventname == "2_year_follow_up_y_arm_1") %>% 
  mutate(youth_sum = case_when(sex == "M" ~ #male
                           (pds_ht2_y + pds_bdyhair_y + pds_skin2_y + pds_m4_y + pds_m5_y),  
                         sex == "F" ~ #female
                           (pds_ht2_y + pds_bdyhair_y + pds_skin2_y + pds_f4_2_y + pds_f5_y))) %>% 
  mutate(youth_mean = youth_sum/5)  %>% 
  select(src_subject_id, eventname, sex, youth_sum, youth_mean)

youth_pds_baseline <- youth_pds_multiwave %>% 
  filter(eventname == "baseline_year_1_arm_1") %>% 
  select(-eventname)
  
youth_pds_followup <- youth_pds_multiwave %>% 
  filter(eventname == "2_year_follow_up_y_arm_1") %>% 
  select(-eventname)
  


#### Parent-report, removes everyone w/o complete data
parent_pds[parent_pds == 777] <- NA
parent_pds[parent_pds == 999] <- NA

#Baseline
parent_pds_multiwave <- parent_pds %>% 
    filter(eventname == "baseline_year_1_arm_1" | eventname == "2_year_follow_up_y_arm_1") %>% 
    mutate(parent_sum = case_when(sex == "M" ~ #male
                           (pds_1_p + pds_2_p + pds_3_p + pds_m4_p + pds_m5_p),  
                         sex == "F" ~ #female
                           (pds_1_p + pds_2_p + pds_3_p + pds_f4_p + pds_f5b_p))) %>% 
  mutate(parent_mean = parent_sum/5) %>% 
  select(src_subject_id, eventname, sex, parent_sum, parent_mean)


parent_pds_baseline <- parent_pds_multiwave %>% 
    filter(eventname == "baseline_year_1_arm_1") %>% 
  select(-eventname)


parent_pds_followup <- parent_pds_multiwave%>% 
    filter(eventname == "2_year_follow_up_y_arm_1")%>% 
  select(-eventname)


#### Cognition

#Baseline
cognition_baseline <- cognition %>% 
  filter(eventname == "baseline_year_1_arm_1") %>% 
  select(-eventname) %>%
  mutate(cog_zscore = scale(nihtbx_totalcomp_uncorrected))

#Follow-Up
cognition_followup <- cognition %>% 
  filter(eventname == "2_year_follow_up_y_arm_1")%>% 
  select(-eventname)

#Multiwave
cognition_multiwave <- cognition %>% 
  filter(eventname == "baseline_year_1_arm_1" | eventname == "2_year_follow_up_y_arm_1")


#Social/GISH
gish[gish== 777] <- NA
gish[gish == 999] <- NA

# sum scores
gish %>% 
  mutate(gishSum = rowSums(across(c(3:15)), na.rm=T))

```
**At this point, switch to the ExistingModel.Rmd file, where you can use renamed_smri_baseline and renamed_smri_followup to make predictions.**

Add age to brainAGE estimates - Exisiting Model
```{r}
#Don't need to run the load lines if you just ran ExistingModel.Rmd, these will already be loaded
load("/Volumes/devbrainlab/Lucy/BrainAGE/fyp/BaselineBrainAgeCorrected.Rda")
load("/Volumes/devbrainlab/Lucy/BrainAGE/fyp/FollowUpBrainAgeCorrected.Rda")

#make renamed copies of predictions
corrected_baseline_brainage <- brain_age_corrected_df_baseline
corrected_followup_brainage <- brain_age_corrected_df_followup

#add subject ids to predictions
corrected_baseline_brainage$src_subject_id <- smri_baseline$src_subject_id

corrected_followup_brainage$src_subject_id <- smri_followup$src_subject_id
```


Split Samples for Analysis
```{r}

brainage_pds_baseline <- corrected_baseline_brainage[, c(6,1:5)] %>% 
  left_join(youth_pds_baseline) %>% 
  left_join(parent_pds_baseline) %>% 
  left_join(cognition_baseline)


#Split sample
smp_size_baseline <- floor(0.5 * nrow(brainage_pds_baseline))

## set the seed to make your partition reproducible
set.seed(123)
sample_ind_baseline <- sample(seq_len(nrow(brainage_pds_baseline)), size = smp_size_baseline)

sample_1_baseline <- brainage_pds_baseline[sample_ind_baseline, ]
sample_2_baseline <- brainage_pds_baseline[-sample_ind_baseline, ]

#t test for age differences
t_apa(t_test(sample_1_baseline$interview_age, sample_2_baseline$interview_age))

#chi-sq for sex differences
#chisq_apa(chisq.test(sample_1_baseline$sex, sample_2_baseline$sex))

prop <- c(length(which(sample_2_baseline$sex == "F")) / length(which(sample_2_baseline$sex == "F" | sample_2_baseline$sex == "M")), length(which(sample_2_baseline$sex == "M")) / length(which(sample_2_baseline$sex == "F" | sample_2_baseline$sex == "M")))

chisq_apa(chisq.test(x = table(sample_1_baseline$sex), p = prop))

#Save samples
save(brainage_pds_baseline, file="BrainagePDSBaseline.Rda")
save(sample_1_baseline, file="Sample1Baseline.Rda")
save(sample_2_baseline, file = "Sample2Baseline.Rda")




# Follow-Up
brainage_pds_followup <- corrected_followup_brainage[, c(6,1:5)] %>% 
  left_join(youth_pds_followup) %>% 
  left_join(parent_pds_followup) %>% 
  left_join(cognition_followup)


#Split sample
smp_size_followup <- floor(0.5 * nrow(brainage_pds_followup))

## set the seed to make your partition reproducible
set.seed(123)
sample_ind_followup <- sample(seq_len(nrow(brainage_pds_followup)), size = smp_size_followup)

sample_1_followup <- brainage_pds_followup[sample_ind_followup, ]
sample_2_followup <- brainage_pds_followup[-sample_ind_followup, ]

#t test for age differences
t_apa(t_test(sample_1_followup$interview_age, sample_2_followup$interview_age))

#chi-sq for sex differences

props <- c(length(which(sample_2_followup$sex == "F")) / length(which(sample_2_followup$sex == "F" | sample_2_followup$sex == "M")), length(which(sample_2_followup$sex == "M")) / length(which(sample_2_followup$sex == "F" | sample_2_followup$sex == "M")))

chisq_apa(chisq.test(x = table(sample_1_followup$sex), p = props))


#Save samples
save(brainage_pds_followup, file="BrainagePDSFollowUp.Rda")
save(sample_1_followup, file="Sample1FollowUp.Rda")
save(sample_2_followup, file = "Sample2FollowUp.Rda")


#Add Gish to follow-up at a later date
```

#################
Make Dataframe for ABCD Model Training & Testing
################
```{r}

#can run the line below if you need to reload smri_abcd_model_baseline
#smri_abcd_model_baseline <- renamed_smri_baseline[, 1:180] 

training_sample_baseline <- filter(smri_abcd_model_baseline, src_subject_id %in% sample_1_baseline$src_subject_id) %>% 
  select(-c(subjectkey, src_subject_id, sex, eventname))

analysis_sample_baseline <- filter(smri_abcd_model_baseline, src_subject_id %in% sample_2_baseline$src_subject_id)

save(training_sample_baseline, file="training_sample_baseline.Rds")
save(analysis_sample_baseline, file="analysis_sample_baseline.Rds")

```


**Switch over to ABCDModel.Rmd to get predictions from ABCD-trained model** 

Attach maturity variables to ABCD model estimates
```{r}
#if you just ran ABCDModel.Rmd, comment out the next line
abcd_brainage_baseline <- rio::import("/Volumes/devbrainlab/Lucy/BrainAGE/fyp/ABCDBaselineBrainAgeCorrected.Rda")
abcd_brainage_baseline$src_subject_id <- sample_2_baseline$src_subject_id

abcd_brainage_baseline <- abcd_brainage_baseline[, c(6,1:5)]


abcd_baseline <- sample_2_baseline %>% 
  select(1, 7:17)

abcd_baseline <- left_join(abcd_baseline, abcd_brainage_baseline)

save(abcd_baseline, file="ABCDBaselineSample.Rda")

#double check that sample 2s remain consistent
abcd_baseline_ids <- abcd_brainage_baseline$src_subject_id

og_sample_2 <- rio::import("/Volumes/devbrainlab/Lucy/BrainAGE/fyp_old/Sample2Baseline.Rda")
og_sample_2_ids <- og_sample_2$src_subject_id

id_compare<-data.frame(abcd_baseline_ids,og_sample_2_ids)
sum(ifelse(id_compare$abcd_baseline_ids==id_compare$og_sample_2_ids,1,0))

#they're the same!


```

**Below section is for future analyses**

#################
Make Dataframe of sMRI features for combined wave model
################
```{r}

#Pull 1800 total
#900 from each
#Pull 900 from baseline
#filter those 900 out of the followup and then pull 900


## set the seed to make your partition reproducible
set.seed(42)
#Pull 900 from baseline
sample_ind_abcd_baseline <- sample(seq_len(nrow(smri_abcd_model_baseline)), size = 900)

abcd_train_baseline <- smri_abcd_model_baseline[sample_ind_abcd_baseline, ]
abcd_test_baseline <- smri_abcd_model_baseline[-sample_ind_abcd_baseline, ]


#filter baseline participants out of follow-up
smri_abcd_model_followup_filtered <- smri_abcd_model_followup %>% 
  filter(!src_subject_id %in% abcd_train_baseline$src_subject_id)

set.seed(42)
#Pull 900 from followup
sample_ind_abcd_followup <- sample(seq_len(nrow(smri_abcd_model_followup_filtered )), size = 900)

abcd_train_followup <- smri_abcd_model_followup[sample_ind_abcd_followup, ]
abcd_test_followup <- smri_abcd_model_followup[-sample_ind_abcd_followup, ]


abcd_train_multiwave <- rbind(abcd_train_baseline, abcd_train_followup) %>% 
  select(-c(subjectkey, src_subject_id, sex, eventname))
save(abcd_train_multiwave, file="abcd_train_multiwave.Rds")

abcd_test_multiwave <- rbind(abcd_test_baseline, abcd_test_followup)
save(abcd_test_multiwave, file="abcd_test_multiwave.Rds")

```


##############
Add Other Variable to Multiwave Test
##############
Add age to brainAGE estimates
```{r}
#load("/Volumes/devbrainlab/Lucy/BrainAGE/fyp/ABCDMultiwaveBrainAgeCorrected.Rda")

corrected_multiwave_brainage <- brain_age_corrected_df_multiwave

corrected_multiwave_brainage$src_subject_id <- abcd_test_multiwave$src_subject_id


#combine waves of youth pds

#combine waves of parent pds

#combine waves of cognition

#change
brainage_pds_multiwave <- corrected_multiwave_brainage[, c(6,1:5)] %>% 
  left_join(youth_pds_multiwave) %>% 
  left_join(parent_pds_multiwave) %>% 
  left_join(cognition_multiwave)


```