Changes

== Pre-Phasing using IMPUTE2 ==

== Pre-Phasing using IMPUTE2 ==

You should now be ready to run the first analysis step, which is to estimate phased haplotypes for each sample. The script below, named ''prototype_phasing_job.sh''(based on the [https://mathgen.stats.ox.ac.uk/impute/prephasing_and_imputation_with_impute2.tgz example script] on the IMPUTE2 website), illustrates how to do this. It requires three parameters (chromosome number, interval start and interval end) and assumes that input data files will in a directory called <code>data_files</code>, with results in <code>results</code> directory and estimated haplotypes in a <code>sampled_haps</code> directory. It has been modified to specify the build 37 recombination map should be used and to include the <code>-allow_large_regions</code> parameter.

You should now be ready to run the first analysis step, which is to estimate phased haplotypes for each sample. The script below, named ''prototype_phasing_job.sh''(based on the [https://mathgen.stats.ox.ac.uk/impute/prephasing_and_imputation_with_impute2.tgz example script] on the IMPUTE2 website), illustrates how to do this. It requires three parameters (chromosome number, interval start and interval end) and assumes that input data files will in a directory called <code>data_files</code>, with results in <code>results</code> directory and estimated haplotypes in a <code>sampled_haps</code> directory. It has been modified to specify the build 37 recombination map should be used and to include the <code>-allow_large_regions</code> parameter and to set the effective population size NE=20,000 (as recommended in the IMPUTE2 website for the more recent 1000 Genome Project builds).

<source lang="bash">

<source lang="bash">

ITER=30

ITER=30

BURNIN=10

BURNIN=10

NE=11500

NE=20000

# reference data files

# reference data files