Create tables of cortical and subcortical measurements
by [Baptiste Couvy-Duchesne] - 23 April 2020 -
1 Extract ROI values from FreeSurfer parcellation
Such tables are not the main ones used in the analyses but they contain ICV and other useful measurements.
They correspond to the ENIGMA - average values used by the ENIGMA consortium (LandRVolumes and ThickAvg / SurfAvg files)
NOTE: this code is identical to that of the ENIGMA processing protocols
http://enigma.ini.usc.edu/protocols/imaging-protocols/.
for batch in {1..5}
do
echo ${batch}
cd ${wd}/FS/FSresults/batch${batch}
echo 'SubjID,L_bankssts_thickavg,L_caudalanteriorcingulate_thickavg,L_caudalmiddlefrontal_thickavg,L_cuneus_thickavg,L_entorhinal_thickavg,L_fusiform_thickavg,L_inferiorparietal_thickavg,L_inferiortemporal_thickavg,L_isthmuscingulate_thickavg,L_lateraloccipital_thickavg,L_lateralorbitofrontal_thickavg,L_lingual_thickavg,L_medialorbitofrontal_thickavg,L_middletemporal_thickavg,L_parahippocampal_thickavg,L_paracentral_thickavg,L_parsopercularis_thickavg,L_parsorbitalis_thickavg,L_parstriangularis_thickavg,L_pericalcarine_thickavg,L_postcentral_thickavg,L_posteriorcingulate_thickavg,L_precentral_thickavg,L_precuneus_thickavg,L_rostralanteriorcingulate_thickavg,L_rostralmiddlefrontal_thickavg,L_superiorfrontal_thickavg,L_superiorparietal_thickavg,L_superiortemporal_thickavg,L_supramarginal_thickavg,L_frontalpole_thickavg,L_temporalpole_thickavg,L_transversetemporal_thickavg,L_insula_thickavg,R_bankssts_thickavg,R_caudalanteriorcingulate_thickavg,R_caudalmiddlefrontal_thickavg,R_cuneus_thickavg,R_entorhinal_thickavg,R_fusiform_thickavg,R_inferiorparietal_thickavg,R_inferiortemporal_thickavg,R_isthmuscingulate_thickavg,R_lateraloccipital_thickavg,R_lateralorbitofrontal_thickavg,R_lingual_thickavg,R_medialorbitofrontal_thickavg,R_middletemporal_thickavg,R_parahippocampal_thickavg,R_paracentral_thickavg,R_parsopercularis_thickavg,R_parsorbitalis_thickavg,R_parstriangularis_thickavg,R_pericalcarine_thickavg,R_postcentral_thickavg,R_posteriorcingulate_thickavg,R_precentral_thickavg,R_precuneus_thickavg,R_rostralanteriorcingulate_thickavg,R_rostralmiddlefrontal_thickavg,R_superiorfrontal_thickavg,R_superiorparietal_thickavg,R_superiortemporal_thickavg,R_supramarginal_thickavg,R_frontalpole_thickavg,R_temporalpole_thickavg,R_transversetemporal_thickavg,R_insula_thickavg,LThickness,RThickness,LSurfArea,RSurfArea,ICV' > ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
echo 'SubjID,L_bankssts_surfavg,L_caudalanteriorcingulate_surfavg,L_caudalmiddlefrontal_surfavg,L_cuneus_surfavg,L_entorhinal_surfavg,L_fusiform_surfavg,L_inferiorparietal_surfavg,L_inferiortemporal_surfavg,L_isthmuscingulate_surfavg,L_lateraloccipital_surfavg,L_lateralorbitofrontal_surfavg,L_lingual_surfavg,L_medialorbitofrontal_surfavg,L_middletemporal_surfavg,L_parahippocampal_surfavg,L_paracentral_surfavg,L_parsopercularis_surfavg,L_parsorbitalis_surfavg,L_parstriangularis_surfavg,L_pericalcarine_surfavg,L_postcentral_surfavg,L_posteriorcingulate_surfavg,L_precentral_surfavg,L_precuneus_surfavg,L_rostralanteriorcingulate_surfavg,L_rostralmiddlefrontal_surfavg,L_superiorfrontal_surfavg,L_superiorparietal_surfavg,L_superiortemporal_surfavg,L_supramarginal_surfavg,L_frontalpole_surfavg,L_temporalpole_surfavg,L_transversetemporal_surfavg,L_insula_surfavg,R_bankssts_surfavg,R_caudalanteriorcingulate_surfavg,R_caudalmiddlefrontal_surfavg,R_cuneus_surfavg,R_entorhinal_surfavg,R_fusiform_surfavg,R_inferiorparietal_surfavg,R_inferiortemporal_surfavg,R_isthmuscingulate_surfavg,R_lateraloccipital_surfavg,R_lateralorbitofrontal_surfavg,R_lingual_surfavg,R_medialorbitofrontal_surfavg,R_middletemporal_surfavg,R_parahippocampal_surfavg,R_paracentral_surfavg,R_parsopercularis_surfavg,R_parsorbitalis_surfavg,R_parstriangularis_surfavg,R_pericalcarine_surfavg,R_postcentral_surfavg,R_posteriorcingulate_surfavg,R_precentral_surfavg,R_precuneus_surfavg,R_rostralanteriorcingulate_surfavg,R_rostralmiddlefrontal_surfavg,R_superiorfrontal_surfavg,R_superiorparietal_surfavg,R_superiortemporal_surfavg,R_supramarginal_surfavg,R_frontalpole_surfavg,R_temporalpole_surfavg,R_transversetemporal_surfavg,R_insula_surfavg,LThickness,RThickness,LSurfArea,RSurfArea,ICV' > ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
for subj_id in $(ls -d *); do
echo ${subj_id}
printf "%s," "${subj_id}" >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%s," "${subj_id}" >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
for side in lh.aparc.stats rh.aparc.stats; do
for x in bankssts caudalanteriorcingulate caudalmiddlefrontal cuneus entorhinal fusiform inferiorparietal inferiortemporal isthmuscingulate lateraloccipital lateralorbitofrontal lingual medialorbitofrontal middletemporal parahippocampal paracentral parsopercularis parsorbitalis parstriangularis pericalcarine postcentral posteriorcingulate precentral precuneus rostralanteriorcingulate rostralmiddlefrontal superiorfrontal superiorparietal superiortemporal supramarginal frontalpole temporalpole transversetemporal insula; do
printf "%g," `grep -w ${x} ${subj_id}/stats/${side} | awk '{print $5}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g," `grep -w ${x} ${subj_id}/stats/${side} | awk '{print $3}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
done
done
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/lh.aparc.stats | grep MeanThickness | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/rh.aparc.stats | grep MeanThickness | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/lh.aparc.stats | grep MeanThickness | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/rh.aparc.stats | grep MeanThickness | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/lh.aparc.stats | grep WhiteSurfArea | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/rh.aparc.stats | grep WhiteSurfArea | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/lh.aparc.stats | grep WhiteSurfArea | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
printf "%g," `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/rh.aparc.stats | grep WhiteSurfArea | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
printf "%g" `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/aseg.stats | grep IntraCranialVol | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
printf "%g" `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/aseg.stats | grep IntraCranialVol | awk -F, '{print $4}'` >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
echo "" >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv
echo "" >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv
done
done2 Check that indivs with missing data were the ones labelled “unusable” by the biobank
3 Extract subcortical volumes
for batch in {1..5}
do
echo ${batch}
cd ${wd}/FS/FSresults/batch${batch}
echo "SubjID,Left-Lateral-Ventricle,Right-Lateral-Ventricle,Left-Thalamus-Proper,Right-Thalamus-Proper,Left-Caudate,Right-Caudate,Left-Putamen,Right-Putamen,Left-Pallidum,Right-Pallidum,Left-Hippocampus,Right-Hippocampus,Left-Amygdala,Right-Amygdala,Left-Accumbens-area,Right-Accumbens-area,Left-Inf-Lat-Vent,Left-Cerebellum-White-Matter,Left-Cerebellum-Cortex,3rd-Ventricle,4th-Ventricle,Brain-Stem,CSF,Left-VentralDC,Left-vessel,Left-choroid-plexus,Right-Inf-Lat-Vent,Right-Cerebellum-White-Matter,Right-Cerebellum-Cortex,Right-VentralDC,Right-vessel,Right-choroid-plexus,WM-hypointensities,Optic-Chiasm,CC_Posterior,CC_Mid_Posterior,CC_Central,CC_Mid_Anterior,CC_Anterior,ICV" > ${wd}/FS/LandRvolumes_batch${batch}.csv
for subj_id in `ls -d *`; do
printf "%s," "${subj_id}" >> ${wd}/FS/LandRvolumes_batch${batch}.csv
for x in Left-Lateral-Ventricle Right-Lateral-Ventricle Left-Thalamus-Proper Right-Thalamus-Proper Left-Caudate Right-Caudate Left-Putamen Right-Putamen Left-Pallidum Right-Pallidum Left-Hippocampus Right-Hippocampus Left-Amygdala Right-Amygdala Left-Accumbens-area Right-Accumbens-area Left-Inf-Lat-Vent Left-Cerebellum-White-Matter Left-Cerebellum-Cortex 3rd-Ventricle 4th-Ventricle Brain-Stem CSF Left-VentralDC Left-vessel Left-choroid-plexus Right-Inf-Lat-Vent Right-Cerebellum-White-Matter Right-Cerebellum-Cortex Right-VentralDC Right-vessel Right-choroid-plexus WM-hypointensities Optic-Chiasm CC_Posterior CC_Mid_Posterior CC_Central CC_Mid_Anterior CC_Anterior; do
printf "%g," `grep ${x} ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/aseg.stats | awk '{print $4}'` >> ${wd}/FS/LandRvolumes_batch${batch}.csv
done
printf "%g" `cat ${wd}/FS/FSresults/batch${batch}/${subj_id}/stats/aseg.stats | grep IntraCranialVol | awk -F, '{print $4}'` >> ${wd}/FS/LandRvolumes_batch${batch}.csv
echo "" >> ${wd}/FS/LandRvolumes_batch${batch}.csv
done
done4 Combine ENIGMA ROI files across batches
# Initialise
cp ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch1.csv ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_UKB.csv
cp ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch1.csv ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_UKB.csv
cp ${wd}/FS/LandRvolumes_batch1.csv ${wd}/FS/LandRvolumes.csv
for batch in {2..5}
do
echo ${batch}
tail -n +2 ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_batch${batch}.csv >> ${wd}/FS/CorticalMeasuresENIGMA_ThickAvg_UKB.csv
tail -n +2 ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_batch${batch}.csv >> ${wd}/FS/CorticalMeasuresENIGMA_SurfAvg_UKB.csv
sed 's/,\+/ /g' ${wd}/FS/LandRvolumes_batch${batch}.csv > ${wd}/FS/LandRvolumes_batch${batch}.txt
sed '1d' ./LandRvolumes_batch${batch}.txt >> ./LandRvolumes.txt
done5 Extract cortical vertex-wise data
Main imaging-derived phenotype tables for the analyses
Here, I got a bit lazy and did not write the piece of code to initialise automatically the vertexnames. Instead I initialise using one individual (batch1/1002XXX) for which I checked processing went well.
for batch in {1..5}
do
for hemi in lh rh
do
for moda in area thickness
do
echo ${hemi}.${moda}
${bind}/qsubshcom " echo \"vertexnum\" > "${wd}"/FS/"${hemi}"."${moda}".batch"${batch}".fwhm0.UKB.txt |;
awk '{print \$1}' "${wd}"/FS/FSresults/batch1/1002XXX/"${hemi}"."${moda}".fwhm0.fsaverage.asc >> "${wd}"/FS/"${hemi}"."${moda}".batch"${batch}".fwhm0.UKB.txt |;
for ID in \$(awk -F\",\" \"NR>0 {print \$1}\" "${medici}"/T1/raw/batch"${batch}"/T1_fetch_batch"${batch}".lis) |;
do |;
ID=\${ID:0:7} |;
echo \${ID} |;
if [ -f "${wd}"/FS/FSresults/batch${batch}/\${ID}/"${hemi}"."${moda}".fwhm0.fsaverage.asc ] |;
then |;
echo \${ID} > "${wd}"/FS/"${hemi}"."${moda}"."${batch}".temp.lta |;
awk '{print \$5}' "${wd}"/FS/FSresults/batch${batch}/\${ID}/"${hemi}"."${moda}".fwhm0.fsaverage.asc >> "${wd}"/FS/"${hemi}"."${moda}"."${batch}".temp.lta |;
paste "${wd}"/FS/"${hemi}"."${moda}".batch"${batch}".fwhm0.UKB.txt "${wd}"/FS/"${hemi}"."${moda}"."${batch}".temp.lta > "${wd}"/FS/"${hemi}"."${moda}"."${batch}".temp2.lta |;
cp "${wd}"/FS/"${hemi}"."${moda}"."${batch}".temp2.lta "${wd}"/FS/"${hemi}"."${moda}".batch"${batch}".fwhm0.UKB.txt |;
fi |; done " 1 4G ExtractData_${batch} 48:00:00 " -acct=UQ-IMB-CNSG"
done
done
done6 Extract subcortical vertex-wise data
Code provided by Boris Gutman as part of the ENIGMA-shape package
# create csv ID list (required input)
for batch in {1..5}
do
cd ${wd}/FS/ENIGMAshapeResults/batch${batch}/
echo "ID,V2" > ${wd}/FS/groupfile_batch${batch}.csv
for d in */ ; do
ID=${d:0:7}
echo ${ID},V2 >> ${wd}/FS/groupfile_batch${batch}.csv
done
done
# Create files
# Replace /working/directory/ by the actual path
for batch in {1..5}
do
# Define variables
groupfile="/working/directory/FS/groupfile_batch"${batch}".csv"
filename=$(basename "${groupfile}")
extension="${filename##*.}"
filename="${filename%.*}"
echo ${groupfile}
# Create left and right datasets
setROIS="10 11 12 13 17 18 26"
${wd}/ENIGMA_shape/MedialDemonsShared/bin/raw_list2CSV_matrix ${wd}/FS/${filename}_LogJacs_lh.csv ${wd}/ENIGMA_shape/MedialDemonsShared/atlas GOF ${setROIS} LogJacs resliced_mesh ${groupfile} ${wd}/FS/ENIGMAshapeResults/batch${batch}
${wd}/ENIGMA_shape/MedialDemonsShared/bin/raw_list2CSV_matrix ${wd}/FS/${filename}_thick_lh.csv ${wd}/ENIGMA_shape/MedialDemonsShared/atlas GOF ${setROIS} thick resliced_mesh ${groupfile} ${wd}/FS/ENIGMAshapeResults/batch${batch}
setROIS="49 50 51 52 53 54 58"
${wd}/ENIGMA_shape/MedialDemonsShared/bin/raw_list2CSV_matrix ${wd}/FS/${filename}_LogJacs_rh.csv ${wd}/ENIGMA_shape/MedialDemonsShared/atlas GOF ${setROIS} LogJacs resliced_mesh ${groupfile} ${wd}/FS/ENIGMAshapeResults/batch${batch}
${wd}/ENIGMA_shape/MedialDemonsShared/bin/raw_list2CSV_matrix ${wd}/FS/${filename}_thick_rh.csv ${wd}/ENIGMA_shape/MedialDemonsShared/atlas GOF ${setROIS} thick resliced_mesh ${groupfile} ${wd}/FS/ENIGMAshapeResults/batch${batch}
# Convert csv to text with space delimited
sed 's/,\+/ /g' ${wd}/FS/${filename}_LogJacs_lh.csv > ${wd}/FS/${filename}_LogJacs_lh.txt
sed 's/,\+/ /g' ${wd}/FS/${filename}_LogJacs_rh.csv > ${wd}/FS/${filename}_LogJacs_rh.txt
sed 's/,\+/ /g' ${wd}/FS/${filename}_thick_lh.csv > ${wd}/FS/${filename}_thick_lh.txt
sed 's/,\+/ /g' ${wd}/FS/${filename}_thick_rh.csv > ${wd}/FS/${filename}_thick_rh.txt
done7 Check dimension of output files (example)
8 Rename vertices and create final text files combining all batches
8.1 Cortical data
First we add the prefixes rht, lht, rha and lha to cortical vertices numbers in order to avoid duplicates when we will put all vertices togeter.
# Initialise
awk 'BEGIN{OFS="\t"}$1="rht_"$1' ${wd}/FS/rh.thickness.batch1.fwhm0.UKB.txt > ${wd}/FS/rh.thickness.fwhm0.UKB.txt
awk 'BEGIN{OFS="\t"}$1="lht_"$1' ${wd}/FS/lh.thickness.batch1.fwhm0.UKB.txt > ${wd}/FS/lh.thickness.fwhm0.UKB.txt
awk 'BEGIN{OFS="\t"}$1="rha_"$1' ${wd}/FS/rh.area.batch1.fwhm0.UKB.txt > ${wd}/FS/rh.area.fwhm0.UKB.txt
awk 'BEGIN{OFS="\t"}$1="lha_"$1' ${wd}/FS/lh.area.batch1.fwhm0.UKB.txt > ${wd}/FS/lh.area.fwhm0.UKB.txt
# Get rid of vertices (row names) and append batch file to the global one
for batch in {2..5}
do
for moda in thickness area
do
for hemi in lh rh
do
echo ${batch}.${hemi}.${moda}.
# Get rid row names (1st column)
awk '{$1=""}1' ${wd}/FS/${hemi}.${moda}.batch${batch}.fwhm0.UKB.txt | awk '{$1=$1}1' > ${wd}/FS/${hemi}.${moda}.batch${batch}.fwhm0.UKB_noColNames.temp
# Paste to the combined file
paste ${wd}/FS/${hemi}.${moda}.fwhm0.UKB.txt ${wd}/FS/${hemi}.${moda}.batch${batch}.fwhm0.UKB_noColNames.temp > ${wd}/FS/${hemi}.${moda}.fwhm0.UKB_2.temp
cp ${wd}/FS/${hemi}.${moda}.fwhm0.UKB_2.temp ${wd}/${hemi}.${moda}.fwhm0.UKB.txt
done
done
done
# Check dimension quickly
awk '{print NF}' ${wd}/FS/lh.area.fwhm0.UKB.txt | sort -nu | tail -n 1
cat ${wd}/FS/lh.area.fwhm0.UKB.txt | wc -l9 Subcortical data
We kept the vertex names used in the ENIGMA-shape package, which are all unique
# Initialise
for moda in thick LogJacs
do
for hemi in lh rh
do
echo ${hemi}.${moda}
cp $wd/FS/groupfile_batch1_${moda}_${hemi}.txt $wd/FS/${hemi}.${moda}.transposed.txt
done
done
# Append
for batch in {2..10}
do
# Get rid of vertices (row names) of files
for moda in thick LogJacs
do
for hemi in lh rh
do
echo ${batch}.${hemi}.${moda}
tail -n +2 ${wd}/FS/groupfile_batch${batch}_${moda}_${hemi}.txt >> ${wd}/FS/${hemi}.${moda}.transposed.txt
done
done
done10 Tidy up, delete files to make room for more processing
# Tar outputs from processing into large tarballs - by batch to faciliate re-use
for batch in {1..5}
do
# Tar and move FS folder
cd ${medici}/UKB_processing
tar -cvf FreeSurferProcessed_FSresults_batch${batch} ${wd}/FS/FSresults/batch${batch}
tar -cvf FreeSurferProcessed_ENIGMAshapeResults_batch${batch} ${wd}/FS/ENIGMAshapeResults/batch${batch}
done
# Example of untarring a single tarball
tar -C /new/working/directory -xvf ${medici}/Archive_FSProcessing/FreeSurferProcessed_batch3A work by by [Baptiste Couvy-Duchesne] - 23 April 2020