nievergeltlab

src='/run/media/genetics/pgc_ptsd/broad/gracy' # Source directory
dst='/run/media/root/_gen/pts1/gracy' # Destination directory


icp() {
  f="${1}";
  [ -d "$f" ] && {
    [ ! -d "${dst}${f#$src}" ] && mkdir -p "${dst}${f#$src}";
    return
  }

nievergeltlab

#List 1000g Europeans
awk '{if ($3 == "EUR") print $1}' integrated_call_samples_v3.20130502.ALL.panel > 1000g_europeans.subjects

for i in 22 # {1..22}
do
 vcf-subset -c 1000g_europeans.subjects ALL.chr"$i".phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz | fill-an-ac | gzip -c > CEU.chr"$i".phase1.vcf.gz
    done

#Get marker info for all markers. Only take rs markers that are di-allelic
for i in {1..22}

nievergeltlab

#Convert all 1000g data to plink binary
for chr in {1..22}
do
 plink2 --vcf 1000g/ALL.chr"$chr".phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz --keep 1000g_eur.ref2 --make-bed --out 1000g/1000geur_chr"$chr"
 plink2 --vcf 1000g/ALL.chr"$chr".phase3_shapeit2_mvncall_integrated_v5a.20130502.genotypes.vcf.gz --keep 1000g_afr.ref2 --make-bed --out 1000g/1000gafr_chr"$chr"
done

nievergeltlab

awk '{print $1"_"$2, $1"_"$2,$6}' pts_mrsc_mix_am-qc.fam > mrsc.pheno 
awk '{print $1"_"$2, $1"_"$2,$4,$5,$6,$7,$8}' pts_mrsc_mix_am-qc-eur_pca.menv.mds_cov > mrsc.cov
awk '{print $1"_"$2, $1"_"$2,$4,$5,$6,$7,$8}' pts_mrsc_mix_am-qc-aam_pca.menv.mds_cov > mrscA.cov

zcat chr1.dose.vcf.gz | head -n 100000 | gzip > chr1.abbr.dose.vcf.gz
/mnt/sdb/genetics/tiff1/hrc_impute/plink --vcf  chr1.dose.vcf.gz --hide-covar --pheno mrsc.pheno --allow-no-sex --covar mrsc.cov --double-id --logistic --ci 0.95 --out mrsc_chr1

#Compare to gwas results
zcat daner_mrsc_eur_analysis_run3.gz | grep -v NA | awk '{if (NR == 1 || $1 == "1") print}' | sort -g -k 11  > chr1_eur.out

nievergeltlab

wd=/home/maihofer/haps/
cd $wd

#Datasets that have (x) or need (y) to rephase haplotype data
 # safr y
 # mrsc y
 # dnhs y
 # gsdc y
 # fscd y
 # nss1 y

nievergeltlab

Case Prep 65	CEPHDNA	Crtl Post 31	Case Prep 29	Ctrl Prep 42	Case Post 64	Case Post 83	Ctrl Prep 75	Case Prep 85	Ctrl Post 81	Ctrl Post 116	Case Prep 107
Case Post 65	Ctrl Prep 179	Ctrl Prep 31	Case Post 29	Ctrl Post 42	Case Prep 64	Case Prep 83	Ctrl Post 75	Case Post 85	Ctrl Prep 81	Ctrl Prep 116	Case Post 107
Ctrl Post 28	Case Prep 72	Ctrl Prep 37	Case Post 93	Case Post 48	Ctrl Prep 56	Case Prep 49	Ctrl Post 66	Ctrl Post 68	Case Prep 95	Case Prep 104	Ctrl Post 136
Ctrl Prep 28	Case Post 72	Ctrl Post 37	Case Prep 93	Case Prep 48	Ctrl Post 56	Case Post 49	Ctrl Prep 66	Ctrl Prep 68	Case Post 95	Case Post 104	Ctrl Prep 136
Ctrl Prep 185	Case Post 90	Case Post 74	Ctrl Prep 39	Case Prep 77	Ctrl Post 60	Ctrl Post 58	Case Prep 87	Ctrl Prep 1	Case Post 97	Case Post 102	Ctrl Prep 73
Ctrl Post 185	Case Prep 90	Case Prep 74	Ctrl Post 39	Case Post 77	Ctrl Prep 60	Ctrl Prep 58	Case Post 87	Ctrl Post 1	Case Prep 97	Case Prep 102	Ctrl Post 73
Case Post 69	Ctrl Prep 35	Case Prep 215	Ctrl Post 41	Ctrl Post 54	Case Prep 78	Ctr

nievergeltlab

#Note according to Katy format: Anyone with value "5-all others" is alphabetically ordered by last name.

#!!! ASSUMING THAT DATA HAS BEEN DOUBLE SORTED BY "Paper order" and "Published Name" !!!
dat_ordered <- read.csv('affiliations_Freeze2.csv',stringsAsFactors=F,header=T)

#Generate single variable for affiliations, easily to work with..
dat_ordered$Afone <- paste(dat_ordered$Institution..EDITED.,dat_ordered$Department..EDITED.,dat_ordered$City,dat_ordered$State,dat_ordered$Country,sep=". ")
dat_ordered$Afone <- sapply(dat_ordered$Afone, gsub, pattern='\xa0', replacement=" " )

nievergeltlab

args <- commandArgs(TRUE)
scriptno <- args[1]

scriptno <- as.numeric(scriptno)
ncore=7 #Assume that node has this many cores

Ngroups=1
fineness=.005
efseq1 <- seq(1.05,1.3,by=fineness)

nievergeltlab

impute_median <- function(x)
{
	med <- median(x,na.rm=T)
	x[is.na(x)] <- med
	return(x)
}

##Read in a .csv file consisting of Barfield CpGs, one row per CpG, one column per sample.
d1 <- read.csv('MRS_Long_barfieldProbes.csv',header=T,nr=20000)
d1$CpG <- d1$X

nievergeltlab

#list all files, make folder for each


ls * | grep .idat | awk 'BEGIN{FS="_"}{print $1}' | sort -u > folderlist.txt

for folders in $(cat folderlist.txt)
do
 mkdir $folders
done