Imputation benchmarks

tasks/phasingUKB/step1_array/step6_run_repbwt.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+for CHR in {1..22}; do
+	VCF=/mnt/project/data/SNParray_hg19_phased_shapeit4/UKB.chr${CHR}.snps_QC_filter.phased_shapeit4.bcf
+	OUT=UKB.chr${CHR}.snps_QC_filter.phased_shapeit4.indices.repbwt
+	TIM=UKB.chr${CHR}.snps_QC_filter.phased_shapeit4.indices.repbwt.time
+	dx run app-swiss-army-knife --folder "/data/SNParray_hg19_phased_shapeit4/repbwt_indices/" -iimage_file="/docker/repbwt_abe107e_20230119.tar.gz" -icmd="/usr/bin/time -vo $TIM repbwt -i ${VCF} -s ${OUT}" --instance-type mem3_ssd1_v2_x8 --priority normal --name repbwt_chr${CHR} -y
+done
+

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/figure/Fig_r2_snp_array.R

@@ -0,0 +1,52 @@
+#DATA LOADING
+library(RColorBrewer)
+COLdiff = brewer.pal(8,"Set1")
+
+CONFMall = read.csv("methods_coding.csv", stringsAsFactors=FALSE)
+
+mets= c("BEAGLE_v5.4","SHAPEIT_v5.0.0")
+refs = c("149,119 WGS samples")
+
+list_legend=""
+list_colours=""
+j=1
+
+	pdf(paste("Fig_r2_snp_array.pdf", sep=""), 6, 6, bg="white")
+	is_new=FALSE
+	par(fig=c(0,1,0,1), new=is_new, mar=c(5,5,4,2)+0.1)
+	plot(1,1,type="n", xlim=c(0.0005,50), ylim=c(0.0,1), ylab=expression(paste("Aggregate ", italic("r"),""^"2"*"")), xlab="British Population Minor Allele Frequency (%)", log="x", xaxt='n', las=1)
+	mtext(paste0("Target: 1000 samples | Reference:", refs[j]), side=3, line=0)
+	axis(1, at=c(0.0001, 0.001, 0.01, 0.1, 1, 10, 50), label=c("0.0001", "0.001","0.01", "0.1", "1", "10", "50"), main=paste0("Imputation performance"))
+	abline(v=c(0.0001,0.0002,0.0005,0.001,0.002,0.005,0.01,0.02,0.05,0.1,0.2,0.5,1,2,5,10,20,50), col="lightgrey", lty=3)
+	abline(h=seq(0,1,0.05), col="lightgrey", lty=3)
+	x0=0
+
+
+
+	listN<-c()
+	listC<-c()
+	#for (i in seq(1, 2, by=1)) 
+	#{
+		SUBmetP <- paste(mets,"_1k",sep="")
+		CONFmetP = CONFMall[which(CONFMall$Prefix %in% SUBmetP),]
+		for (r in 1:nrow(CONFmetP)) 
+		{
+		  file=paste("data/imputed_1k_rp_",CONFmetP$Met1[r],"_chr20.rsquare.grp.txt.gz", sep="")
+		  print(file)
+		  if(!file.exists(file))
+		  {
+			next
+		  }
+		  D = read.table(file, head=FALSE, stringsAsFactors=FALSE)
+		  points(D$V3*100, D$V5, type="o", pch=20, col=COLdiff[r], lwd=2)
+
+		}
+		listN<-c(listN, paste0(CONFmetP$Met, "-phased reference panel"))
+		listC<-c(listC, COLdiff[1:3])
+	#}
+	print(listN)
+	print(listC)
+	legend("bottomright", legend=listN, fill=listC, bg="white",cex = 0.8)
+
+	dev.off()
+#}

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/figure/convert2png.sh

@@ -0,0 +1,5 @@
+#!/bin/bash
+
+for FILE in $(find . -name "*.pdf"); do
+	convert -quality 100 -density 100 $FILE $(dirname $FILE)/$(basename $FILE .pdf)\.png
+done

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/figure/methods_coding.csv

@@ -0,0 +1,3 @@
+Prefix,Cov,ImpFolder,NameRef,NameRef2,NameTar,Met,Met1,Vtype,Chr,Label,Version,Hex,Hex2,lty,ch
+BEAGLE_v5.4_1k,UKB axiom array,imputation,UKB WGS - Beagle v5.4 phased,UK Biobank WGS (140k samples),100,BEAGLE v5.4,beagle5.4,all,20,-,v5.4,#d40025,#d40025,1,20
+SHAPEIT_v5.0.0_1k,UKB axiom array,imputation,UKB WGS - SHAPEIT v5.0.0 phased,UK Biobank WGS (140k samples),100,SHAPEIT v5.0.0,shapeit5,all,20,-,v5.0.0,#ff8095,#E69F00,1,20

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/step1_run_snparray_phasing_and_imp.sh

@@ -0,0 +1,32 @@
+#!/bin/bash
+
+#Setting 1
+PATHB="Phasing/PhasingWGS/step3_runbeagle/N147754"
+EXTB="vcf.gz"
+METB="beagle5.4"
+
+#Setting 2
+PATHS="Phasing/PhasingWGS/step5_runshapeit_phase2/N147754"
+EXTS="default.bcf"
+METS="shapeit5"
+
+for CHR in 20; do
+		#Setting 2PhasingWGS_Official_release/step2_phase_rare/concat_benchmark/
+		#JOBID0=$(dx run app-swiss-army-knife -iin="/Phasing/PhasingWGS/step8_imputation/support/1k_samples_wbi.txt" -icmd="bcftools view /mnt/project/PhasingWGS_Official_release/step3_phase_beagle/UKB_chr${CHR}.full.${METB}.vcf.gz -r chr${CHR} -S ^1k_samples_wbi.txt -Ou  --threads 4 | bcftools filter -e 'AC==0 || AC==AN' -Oz -o UKB_chr${CHR}.rp.${METB}.vcf.gz --threads 4 && bcftools index -f UKB_chr${CHR}.rp.${METB}.vcf.gz --threads 4" --folder="/PhasingWGS_Official_release/step3_phase_beagle/reference_panels" --instance-type mem1_ssd1_v2_x4 --priority low --name subset_${METB}_chr${CHR} -y | tail -n1 | cut -d" " -f3) 
+
+		#JOBID1=$(dx run app-swiss-army-knife -iin="/docker/bref3.22Jul22.46e.jar" -icmd="java -Xmx32G -jar bref3.22Jul22.46e.jar /mnt/project/PhasingWGS_Official_release/step3_phase_beagle/reference_panels/UKB_chr${CHR}.rp.${METB}.vcf.gz > UKB_chr${CHR}.rp.${METB}.bref3" --folder="/PhasingWGS_Official_release/step3_phase_beagle/reference_panels" --instance-type mem3_ssd1_v2_x4 --priority high --name bref_${METB}_chr${CHR} -y | tail -n1 | cut -d" " -f3) #--depends-on ${JOBID0}
+
+		#from vcf.gz to bref3	
+		#JOBID2=$(dx run app-swiss-army-knife -iin="/docker/beagle.22Jul22.46e.jar" -icmd="java -Xmx32G -jar beagle.22Jul22.46e.jar gt=/mnt/project/Phasing/PhasingWGS/step8_imputation/target_snp_array/axiom_1k_c${CHR}_b0_v2.b38.vcf.gz map=/mnt/project/data/plink_maps/plink.prefix.chr${CHR}.GRCh38.map ref=/mnt/project/PhasingWGS_Official_release/step3_phase_beagle/reference_panels/UKB_chr${CHR}.rp.${METB}.bref3 out=imputed_1k_rp_${METB}_chr${CHR} nthreads=4 chrom=chr${CHR} impute=true gp=true && bcftools index -f imputed_1k_rp_${METB}_chr${CHR}.vcf.gz --threads 4" --folder="/PhasingWGS_Official_release/step3_phase_beagle/imputation" --instance-type mem3_ssd1_v2_x4 --priority low --name imp_rp_${METB}_chr${CHR} -y --depends-on ${JOBID1} | tail -n1 | cut -d" " -f3)
+
+		####################################
+
+		#Setting 2
+		JOBID0=$(dx run app-swiss-army-knife -iin="/Phasing/PhasingWGS/step8_imputation/support/1k_samples_wbi.txt" -icmd="bcftools view /mnt/project/PhasingWGS_Official_release/step2_phase_rare/concat_benchmark/chr${CHR}/UKB_chr${CHR}.full.${METS}_rare.bcf -r chr${CHR} -S ^1k_samples_wbi.txt -Ou  --threads 4 | bcftools filter -e 'AC==0 || AC==AN' -Oz -o UKB_chr${CHR}.rp.${METS}.vcf.gz --threads 4 && bcftools index -f UKB_chr${CHR}.rp.${METS}.vcf.gz --threads 4" --folder="/PhasingWGS_Official_release/step2_phase_rare/reference_panels" --instance-type mem1_ssd1_v2_x4 --priority low --name subset_${METS}_chr${CHR} -y | tail -n1 | cut -d" " -f3) 
+
+		JOBID1=$(dx run app-swiss-army-knife -iin="/docker/bref3.22Jul22.46e.jar" -icmd="java -Xmx32G -jar bref3.22Jul22.46e.jar /mnt/project/PhasingWGS_Official_release/step2_phase_rare/reference_panels/UKB_chr${CHR}.rp.${METS}.vcf.gz > UKB_chr${CHR}.rp.${METS}.bref3" --folder="/PhasingWGS_Official_release/step2_phase_rare/reference_panels" --instance-type mem3_ssd1_v2_x4 --depends-on ${JOBID0} --priority high --name bref_${METS}_chr${CHR} -y | tail -n1 | cut -d" " -f3) #--depends-on ${JOBID0}
+
+		#from vcf.gz to bref3	
+		JOBID2=$(dx run app-swiss-army-knife -iin="/docker/beagle.22Jul22.46e.jar" -icmd="java -Xmx32G -jar beagle.22Jul22.46e.jar gt=/mnt/project/Phasing/PhasingWGS/step8_imputation/target_snp_array/axiom_1k_c${CHR}_b0_v2.b38.vcf.gz map=/mnt/project/data/plink_maps/plink.prefix.chr${CHR}.GRCh38.map ref=/mnt/project/PhasingWGS_Official_release/step2_phase_rare/reference_panels/UKB_chr${CHR}.rp.${METS}.bref3 out=imputed_1k_rp_${METS}_chr${CHR} nthreads=4 chrom=chr${CHR} impute=true gp=true && bcftools index -f imputed_1k_rp_${METS}_chr${CHR}.vcf.gz --threads 4" --folder="/PhasingWGS_Official_release/step2_phase_rare/imputation" --instance-type mem3_ssd1_v2_x4 --priority low --name imp_rp_${METS}_chr${CHR} -y --depends-on ${JOBID1} | tail -n1 | cut -d" " -f3)
+
+done

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/step2_ligate_snp_array.sh

@@ -0,0 +1,13 @@
+#!/bin/bash
+
+METB="beagle5.4"
+METS="shapeit5"
+
+for CHR in 20; do
+	OFOLDER=/Phasing/PhasingWGS/step8_imputation/ligated
+	dx run app-swiss-army-knife -iimage_file="docker/shapeit_ligate_0.0.1.tar.gz" -icmd="ls -1v /mnt/project/Phasing/PhasingWGS/step8_imputation/imputation/${METB}/imputed_1k_rp_${METB}_chr${CHR}_*.vcf.gz > list.txt && shapeit_ligate_v0.0.1 --input list.txt --output imputed_1k_rp_${METB}_chr${CHR}.bcf --index --thread 2 && bcftools index -f imputed_1k_rp_${METB}_chr${CHR}.bcf --threads 2 && rm list.txt" --instance-type mem2_ssd1_v2_x2 --priority low --name ${METB}_lg_${CHR} --folder "${OFOLDER}" -y
+
+	OFOLDER=/Phasing/PhasingWGS/step8_imputation/ligated
+	dx run app-swiss-army-knife -iimage_file="docker/shapeit_ligate_0.0.1.tar.gz" -icmd="ls -1v /mnt/project/Phasing/PhasingWGS/step8_imputation/imputation/${METS}/imputed_1k_rp_${METS}_chr${CHR}_*.vcf.gz > list.txt && shapeit_ligate_v0.0.1 --input list.txt --output imputed_1k_rp_${METS}_chr${CHR}.bcf --index --thread 2 && bcftools index -f imputed_1k_rp_${METS}_chr${CHR}.bcf --threads 2 && rm list.txt" --instance-type mem2_ssd1_v2_x2 --priority low --name ${METS}_lg_${CHR} --folder "${OFOLDER}" -y
+done
+

tasks/phasingUKB/step2_wgs/step10_chr20_imputation_no_chunks/step3_run_conc_cluster.sh

@@ -0,0 +1,35 @@
+#!/bin/bash
+
+#FULL CHR
+OFOLDER=/Phasing/PhasingWGS/step8_imputation/concordance_test/binning2
+
+METB="beagle5.4"
+METS="shapeit5"
+
+########### OLD
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20220825_502047a.tar.gz" -icmd="echo \"chr20 /mnt/project/data/ukb_wgs/unphased/step1_wbic_noval/ukb23352_c20_qc_v1_wbic_no10kgreml_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METB}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --allele-counts --thread 2 --input concordance.txt --output imputed_1k_rp_${METB}_chr${CHR} && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METB}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20220825_502047a.tar.gz" -icmd="echo \"chr20 /mnt/project/data/ukb_wgs/unphased/step1_wbic_noval/ukb23352_c20_qc_v1_wbic_no10kgreml_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METS}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --allele-counts --thread 2 --input concordance.txt --output imputed_1k_rp_${METS}_chr${CHR} && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METS}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221002_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/rp_s1k_shapeit5_chr20.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METB}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --allele-counts --thread 2 --input concordance.txt --output imputed_1k_rp_${METB}_chr${CHR}_rp_af && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METB}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221002_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/rp_s1k_shapeit5_chr20.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METS}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --allele-counts --thread 2 --input concordance.txt --output imputed_1k_rp_${METS}_chr${CHR}_rp_af && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METS}_conc_wgs --folder "/Phasing/PhasingWGS/step8_imputation/concordance_test/" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221005_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/data/ukb_wgs/unphased/step1_wbic_noval/ukb23352_c20_qc_v1_wbic_no10kgreml_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METB}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000  2000 5000 10000  20000 50000 100000 146754 --input concordance.txt --output imputed_1k_rp_${METB}_chr${CHR}_wbi_af_binning2 && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METB}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221005_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/data/ukb_wgs/unphased/step1_wbic_noval/ukb23352_c20_qc_v1_wbic_no10kgreml_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METS}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000  2000 5000 10000  20000 50000 100000 146754 --thread 2 --input concordance.txt --output imputed_1k_rp_${METS}_chr${CHR}_wbi_af_binning2 && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METS}_conc_wgs --folder "${OFOLDER}" -y
+
+
+##########BINNING 2 RP - NEW, for the paper
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221005_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/rp_s1k_shapeit5_chr20.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METB}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000  2000 5000 10000  20000 50000 100000 146754 --thread 2 --input concordance.txt --output imputed_1k_rp_${METB}_chr${CHR}_rp_af_binning2 && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METB}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221005_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/rp_s1k_shapeit5_chr20.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_${METS}_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000  2000 5000 10000  20000 50000 100000 146754 --thread 2 --input concordance.txt --output imputed_1k_rp_${METS}_chr${CHR}_rp_af_binning2 && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name ${METS}_conc_wgs --folder "${OFOLDER}" -y
+
+	#dx run app-swiss-army-knife -iimage_file="/docker/glimpse2_20221005_73decec-dirty.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/rp_s1k_shapeit5_chr20.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/Phasing/PhasingWGS/step8_imputation/ligated/imputed_1k_rp_hrc_chr${CHR}.bcf\" > concordance.txt && GLIMPSE_concordance_v2.0.0 --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000  2000 5000 10000  20000 50000 100000 146754 --thread 2 --input concordance.txt --output imputed_1k_rp_hrc_chr${CHR}_rp_af_binning2 && rm -f concordance.txt" --instance-type mem1_ssd1_v2_x2 --priority low --name hrc_conc_wgs --folder "${OFOLDER}" -y
+
+
+	JOBID2=$(dx run app-swiss-army-knife -iimage_file="/docker/glimpse_v2.0.0-27-g0919952_20221207.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/chr20/rp_ukb23352_c20_qc_v1_all_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/PhasingWGS_Official_release/step2_phase_rare/imputation/imputed_1k_rp_shapeit5_chr20.vcf.gz\" > concordance.txt && GLIMPSE2_concordance --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000 2000 5000 10000  20000 50000 100000 149754 --thread 4 --input concordance.txt --output imputed_1k_rp_shapeit5_chr20 && rm -f concordance.txt" --instance-type mem2_ssd1_v2_x4 --priority low --name cnc_rp_shapeit5 --folder "/PhasingWGS_Official_release/step2_phase_rare/concordance/" -y | tail -n1 | cut -d" " -f3)
+
+
+	JOBID2=$(dx run app-swiss-army-knife -iimage_file="/docker/glimpse_v2.0.0-27-g0919952_20221207.tar.gz" -icmd="echo \"chr20 /mnt/project/Phasing/PhasingWGS/step8_imputation/reference_panels/sites/chr20/rp_ukb23352_c20_qc_v1_all_sites.bcf /mnt/project/data/ukb_wgs/phased_shapeit/panels/validation_greml/wgs_ukb23352_c20_qc_v1_all.bcf /mnt/project/PhasingWGS_Official_release/step3_phase_beagle/imputation/imputed_1k_rp_beagle5.4_chr20.vcf.gz\" > concordance.txt && GLIMPSE2_concordance --gt-val --ac-bins 1 5 10 20 50 100 200 500 1000 2000 5000 10000  20000 50000 100000 149754 --thread 4 --input concordance.txt --output imputed_1k_rp_beagle5.4_chr20 && rm -f concordance.txt" --instance-type mem2_ssd1_v2_x4 --priority low --name cnc_rp_beagle5.4 --folder "/PhasingWGS_Official_release/step3_phase_beagle/concordance/" -y | tail -n1 | cut -d" " -f3)
+

tasks/phasingUKB/step2_wgs/step9_imputation_rerun_new_chunks/data/new.chunks.rerun.size4Mb.txt

@@ -0,0 +1,3 @@
+6	chr1	chr1:107388953-152564670	chr1:108032695-150130228
+7	chr1	chr1:149517389-168622808	chr1:150130247-168122803
+3	chr9	chr9:31836957-77518833	chr9:32674876-76851846

tasks/phasingUKB/step2_wgs/step9_imputation_rerun_new_chunks/data/new.chunks.rerun.size4Mb_2.txt

@@ -0,0 +1,4 @@
+34	chr1	chr1:119707834-146165488	chr1:119957834-145913683	25955850	757198
+35	chr1	chr1:145663679-150599883	chr1:145913685-150349882	4436198	757196
+36	chr1	chr1:150099876-154102132	chr1:150349888-153852121	3502234	757200
+12	chr9	chr9:38478780-67628670	chr9:38729039-67378125	28649087	1262471	28284

tasks/phasingUKB/step2_wgs/step9_imputation_rerun_new_chunks/figure/Fig_r2_snp_array.R

@@ -0,0 +1,52 @@
+#DATA LOADING
+library(RColorBrewer)
+COLdiff = brewer.pal(8,"Set1")
+
+CONFMall = read.csv("methods_coding.csv", stringsAsFactors=FALSE)
+
+mets= c("BEAGLE_v5.4","SHAPEIT_v5.0.0")
+refs = c("140,119 WGS samples")
+
+list_legend=""
+list_colours=""
+j=1
+
+	pdf(paste("Fig_r2_snp_array.pdf", sep=""), 6, 6, bg="white")
+	is_new=FALSE
+	par(fig=c(0,1,0,1), new=is_new, mar=c(5,5,4,2)+0.1)
+	plot(1,1,type="n", xlim=c(0.0005,50), ylim=c(0.0,1), ylab=expression(paste("Aggregate ", italic("r"),""^"2"*"")), xlab="British Population Minor Allele Frequency (%)", log="x", xaxt='n', las=1)
+	mtext(paste0("Target: 1000 samples | Reference:", refs[j]), side=3, line=0)
+	axis(1, at=c(0.0001, 0.001, 0.01, 0.1, 1, 10, 50), label=c("0.0001", "0.001","0.01", "0.1", "1", "10", "50"), main=paste0("Imputation performance"))
+	abline(v=c(0.0001,0.0002,0.0005,0.001,0.002,0.005,0.01,0.02,0.05,0.1,0.2,0.5,1,2,5,10,20,50), col="lightgrey", lty=3)
+	abline(h=seq(0,1,0.05), col="lightgrey", lty=3)
+	x0=0
+
+
+
+	listN<-c()
+	listC<-c()
+	#for (i in seq(1, 2, by=1)) 
+	#{
+		SUBmetP <- paste(mets,"_1k",sep="")
+		CONFmetP = CONFMall[which(CONFMall$Prefix %in% SUBmetP),]
+		for (r in 1:nrow(CONFmetP)) 
+		{
+		  file=paste("data/imputed_1k_rp_",CONFmetP$Met1[r],"_chr20.rsquare.grp.txt.gz", sep="")
+		  print(file)
+		  if(!file.exists(file))
+		  {
+			next
+		  }
+		  D = read.table(file, head=FALSE, stringsAsFactors=FALSE)
+		  points(D$V3*100, D$V5, type="o", pch=20, col=COLdiff[r], lwd=2)
+
+		}
+		listN<-c(listN, paste0(CONFmetP$Met, "-phased reference panel"))
+		listC<-c(listC, COLdiff[1:3])
+	#}
+	print(listN)
+	print(listC)
+	legend("bottomright", legend=listN, fill=listC, bg="white",cex = 0.8)
+
+	dev.off()
+#}