Pair Matcher (PaM): a computer model to optimise pairings using demographic and genetic data
Pre-requisites:
-
linux operating system such as Ubuntu
-
R programming language (and R studio or RKward) installed
Running the package:
-
place uncompressed PaM_local folder on C drive
-
copy plink files (bed/bim/fam) containing 'your' cohort snp data to
PaM_local/uploaded_data folder
and name these files as:
plink.bed plink.bim plink.fam
-
in an Xterm window cd PaM_local
-
ensure linux script is executable using:
chmod +x elhaik_pp.sh
- launch script using
./elhaik_pp.sh
8a) results folder contains the pairs/unpaired in: pairs_case_only.csv & unpaired_individuals.csv
8b) results folder also contains the admixture components for all individuals & matcher run/screen dump as: merge.9.Q & matcher.Rout
-
test folder contains the input files & results for a small data test (13 individuals)
-
note that run time on i5 laptop using 3 cores for ~3900 individuals each containing ~115k SNPs is ~7 hours