Data Processing

First steps

A list of individuals present in the local ancestry calls downloaded from 1KGP was first compiled. From the directory where the ASW archive was extracted, the trio-phased BED files are unpacked and the individual identifiers are listed (one per line):

tar -xzf ./ASW/TrioPhased/bed_files/ASW_bed.tar.gz
ls ./ASW/TrioPhased/*_A_final.bed | xargs -n 1 basename | cut -d _ -f 1 \
> ./ASW/TrioPhased/individuals.txt

Sex information for individuals was obtained from the file:

integrated_call_samples_v3.20130502.ALL.panel

available at:

https://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/.

We then identified individuals present in the local ancestry data but missing from this panel.

Individuals Not Included in the Panel

The following command identifies individuals present in the ancestry calls but absent from the panel:

awk '
NR==FNR { seen[$1]=1; next }
!($1 in seen) { print $1 }
' integrated_call_samples_v3.20130502.ALL.panel ./ASW/TrioPhased/individuals.txt

Output:

NA19985
NA20322
NA20336
NA20341
NA20344

Identification of Related Individuals

These individuals correspond to related samples, as listed in:

20140625_related_individuals.txt

available at:

https://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/.

The following command confirms this:

awk '
NR==FNR { seen[$1]=1; next }
($1 in seen) { print $1 }
' 20140625_related_individuals.txt ./ASW/TrioPhased/individuals.txt

Output:

NA19985
NA20322
NA20336
NA20341
NA20344

Filtering Related Individuals

These individuals are present in the local ancestry calls but are not independent. They should therefore be removed.

To exclude related individuals:

awk '
NR==FNR { seen[$1]=1; next }
!($1 in seen) { print $1 }
' 20140625_related_individuals.txt ./ASW/TrioPhased/individuals.txt

To save the filtered list:

awk '
NR==FNR { seen[$1]=1; next }
!($1 in seen) { print $1 }
' 20140625_related_individuals.txt ./ASW/TrioPhased/individuals.txt \
> ./ASW/TrioPhased/individuals_unrelated.txt

Output Formatting

To format the list of unrelated individuals as a CSV string (e.g., for inclusion in a driver file):

awk '
NR==FNR { seen[$1]=1; next }
!($1 in seen) { printf "\"%s\"," , $1 }
' 20140625_related_individuals.txt ./ASW/TrioPhased/individuals.txt

Male Individuals

To extract the male individuals among the unrelated ones, in the same format, using the sex (gender column) given in the panel file:

awk '
NR==FNR { if ($4 == "male") male[$1]=1; next }
($1 in male) { printf "\"%s\"," , $1 }
' integrated_call_samples_v3.20130502.ALL.panel ./ASW/TrioPhased/individuals_unrelated.txt

These two lists correspond to the fields individual_names and male_names of the driver files used in the ASW examples (61 unrelated individuals, of which 26 are males).