def bi_allelic_site_inbreeding_expr(
call: hl.expr.CallExpression,
) -> hl.expr.Float32Expression:
"""
Return the site inbreeding coefficient as an expression to be computed on a MatrixTable.
This is implemented based on the GATK InbreedingCoeff metric:
https://software.broadinstitute.org/gatk/documentation/article.php?id=8032
.. note::
The computation is run based on the counts of alternate alleles and thus should only be run on bi-allelic sites.
:param call: Expression giving the calls in the MT
:return: Site inbreeding coefficient expression
"""
def inbreeding_coeff(
gt_counts: hl.expr.DictExpression,
) -> hl.expr.Float32Expression:
n = gt_counts.get(0, 0) + gt_counts.get(1, 0) + gt_counts.get(2, 0)
p = (2 * gt_counts.get(0, 0) + gt_counts.get(1, 0)) / (2 * n)
q = (2 * gt_counts.get(2, 0) + gt_counts.get(1, 0)) / (2 * n)
return 1 - (gt_counts.get(1, 0) / (2 * p * q * n))
return hl.bind(inbreeding_coeff, hl.agg.counter(call.n_alt_alleles()))
def _ac_an_parent_child_count(
proband_gt: hl.expr.CallExpression,
father_gt: hl.expr.CallExpression,
mother_gt: hl.expr.CallExpression,
) -> Dict[str, hl.expr.Int64Expression]:
"""Get AC and AN for parents and children."""
ac_parent_expr = hl.agg.sum(father_gt.n_alt_alleles() +
mother_gt.n_alt_alleles())
an_parent_expr = hl.agg.sum(
(hl.is_defined(father_gt) + hl.is_defined(mother_gt)) * 2)
ac_child_expr = hl.agg.sum(proband_gt.n_alt_alleles())
an_child_expr = hl.agg.sum(hl.is_defined(proband_gt) * 2)
return {
f"ac_parents": ac_parent_expr,
f"an_parents": an_parent_expr,
f"ac_children": ac_child_expr,
f"an_children": an_child_expr,
}
