Skip to content

Restructure imputed data

After the imputation process, the data must be stratified by the five superpopulations (EUR, EAS, AMR, AFR, SAS) to enable population-specific evaluation.

Requirements

  • Ubuntu 22.04 (8 CPUs, 32 GB)
  • bcftools (version==1.13)

Low-pass sequencing data

Input data

Merge imputed data

Code

Merge samples from imputed batches 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
set -ue

## Input
CHR=$1
IMPUTED_FOLDER=$2
LPS_COV=$3
TOTAL_SAMPLE=$4

## Merge batches
merge_batches.sh ${IMPUTED_FOLDER} "chr${CHR}_${LPS_COV}" "tem.vcf.gz"
rename_samples.sh tem.vcf.gz "_${LPS_COV}" chr${CHR}_${LPS_COV}_merged_all.vcf.gz ${TOTAL_SAMPLE}
rm tem.vcf.gz

Restruct imputed LPS VCFs

Code

Imputed VCFs is restructured by supperpopulation 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
set -ue

POPULATION_META=$1
POP_NAME=$2
MERGED_VCF=$3
CHR=$4
LPS_COV=$5

## Get sample list for the specified population
awk -F'\t' -v pop_name=${POP_NAME} 'NR!=1 && $6==pop_name {print $1}' ${POPULATION_META} > ${POP_NAME}_sample_list.txt

# Filter the merged VCF for the specified population
bcftools view -S ${POP_NAME}_sample_list.txt ${MERGED_VCF} | bgzip > chr${CHR}_${LPS_COV}_${POP_NAME}_imputed.vcf.gz

Output data

  • restructed lpWGS VCFs

Pseudo SNP Arrays data

Input data

Merge imputed data

Code

1
2
3
4
5
6
7
8
set -ue

## Input
CHR=$1
IMPUTED_FOLDER=$2
LPS_COV=$3

merge_array_batches.sh ${IMPUTED_FOLDER} "${LPS_COV}_chr${CHR}" "chr${CHR}_${LPS_COV}_merged_all.vcf.gz"

Restruct imputed Pseudo-array VCFs

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
set -ue

POPULATION_META=$1
POP_NAME=$2
MERGED_VCF=$3
CHR=$4
LPS_COV=$5

## Get sample list for the specified population
awk -F'\t' -v pop_name=${POP_NAME} 'NR!=1 && $6==pop_name {print $1}' ${POPULATION_META} > ${POP_NAME}_sample_list.txt

## Filter the merged VCF for the specified population
bcftools view -S ${POP_NAME}_sample_list.txt ${MERGED_VCF} | bgzip > chr${CHR}_${LPS_COV}_${POP_NAME}_imputed.vcf.gz

Output data

  • restructed SNP-array VCFs

Prepare true VCFs according supperpopulation

Preparation of true VCFs involves extracting the corresponding samples from the reference panel for each of the five superpopulations.

Input data

Processing

Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
set -ue

## Input
POPULATION_META=$1
POP_NAME=$2
TRUE_VCF=$3
CHR=$4

## Get sample list for the specified population
awk -F'\t' -v pop_name=${POP_NAME} 'NR!=1 && $6==pop_name {print $1}' ${POPULATION_META} > ${POP_NAME}_sample_list.txt

## VCF_true_population_slipt
bcftools view -S ${POP_NAME}_sample_list.txt ${TRUE_VCF} | bgzip > chr${CHR}_${POP_NAME}_true.vcf.gz

Output data

  • True VCFs being collected by supperpopulation