Change wording, fix pc_relate signature

Author: ravwojdyla

sgkit/stats/aggregation.py

@@ -256,10 +256,10 @@ def variant_stats(
         :func:`sgkit.create_genotype_call_dataset`.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     call_genotype_mask
         Input variable name holding call_genotype_mask.
-        As defined by :data:`sgkit.variables.call_genotype_mask`
+        Defined by :data:`sgkit.variables.call_genotype_mask`
     variant_allele_count
         Optional name of the input variable holding variant_allele_count,
         as defined by :data:`sgkit.variables.variant_allele_count`.

sgkit/stats/association.py

@@ -138,13 +138,13 @@ def gwas_linear_regression(
         Dataset containing necessary dependent and independent variables.
     dosage
         Name of genetic dosage variable.
-        As defined by :data:`sgkit.variables.dosage`.
+        Defined by :data:`sgkit.variables.dosage`.
     covariates
         Names of covariate variables (1D or 2D).
-        As defined by :data:`sgkit.variables.covariates`.
+        Defined by :data:`sgkit.variables.covariates`.
     traits
         Names of trait variables (1D or 2D).
-        As defined by :data:`sgkit.variables.traits`.
+        Defined by :data:`sgkit.variables.traits`.
     add_intercept
         Add intercept term to covariate set, by default True.
     merge

sgkit/stats/hwe.py

@@ -146,10 +146,10 @@ def hardy_weinberg_test(
         (in that order) across all samples for a variant.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     call_genotype_mask
         Input variable name holding call_genotype_mask.
-        As defined by :data:`sgkit.variables.call_genotype_mask`
+        Defined by :data:`sgkit.variables.call_genotype_mask`
     merge
         If True (the default), merge the input dataset and the computed
         output variables into a single dataset, otherwise return only

sgkit/stats/pc_relate.py

@@ -34,8 +34,8 @@ def _collapse_ploidy(
 
 def pc_relate(
     ds: xr.Dataset,
-    maf: float = 0.01,
     *,
+    maf: float = 0.01,
     call_genotype: str = "call_genotype",
     call_genotype_mask: str = "call_genotype_mask",
     sample_pcs: str = "sample_pcs",
@@ -82,13 +82,13 @@ def pc_relate(
         The default value is 0.01. Must be between (0.0, 0.1).
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     call_genotype_mask
         Input variable name holding call_genotype_mask.
-        As defined by :data:`sgkit.variables.call_genotype_mask`
+        Defined by :data:`sgkit.variables.call_genotype_mask`
     sample_pcs
         Input variable name holding sample_pcs.
-        As defined by :data:`sgkit.variables.sample_pcs`
+        Defined by :data:`sgkit.variables.sample_pcs`
     merge
         If True (the default), merge the input dataset and the computed
         output variables into a single dataset, otherwise return only

sgkit/stats/popgen.py

@@ -30,7 +30,7 @@ def diversity(
         Genotype call dataset.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     allele_counts
         allele counts to use or calculate, as defined by
         :data:`sgkit.variables.variant_allele_count`
@@ -73,7 +73,7 @@ def divergence(
         Genotype call dataset.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     allele_counts
         allele counts to use or calculate, as defined by
         :data:`sgkit.variables.variant_allele_count`
@@ -123,7 +123,7 @@ def Fst(
         Genotype call dataset.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     allele_counts
         allele counts to use or calculate, as defined by
         :data:`sgkit.variables.variant_allele_count`
@@ -155,15 +155,14 @@ def Tajimas_D(
         Genotype call dataset.
     call_genotype
         Input variable name holding call_genotype.
-        As defined by :data:`sgkit.variables.call_genotype`.
+        Defined by :data:`sgkit.variables.call_genotype`.
     allele_counts
         allele counts to use or calculate, as defined by
         :data:`sgkit.variables.variant_allele_count`
 
     Returns
     -------
     Tajimas' D value.
-
     """
     if allele_counts not in ds:
         variables.validate(ds, {call_genotype: variables.call_genotype})

sgkit/stats/regenie.py

@@ -755,16 +755,16 @@ def regenie(
     ----------
     dosage
         Name of genetic dosage variable.
-        As defined by :data:`sgkit.variables.dosage`.
+        Defined by :data:`sgkit.variables.dosage`.
     covariates
         Names of covariate variables (1D or 2D).
-        As defined by :data:`sgkit.variables.covariates`.
+        Defined by :data:`sgkit.variables.covariates`.
     traits
         Names of trait variables (1D or 2D).
-        As defined by :data:`sgkit.variables.traits`.
+        Defined by :data:`sgkit.variables.traits`.
     variant_contig
         Name of the variant contig input variable.
-        As definied by :data:`sgkit.variables.variant_contig`.
+        Definied by :data:`sgkit.variables.variant_contig`.
     variant_block_size
         Number of variants in each block.
         If int, this describes the number of variants in each block
@@ -808,17 +808,17 @@ def regenie(
     A dataset containing the following variables:
 
     - `base_prediction` (blocks, alphas, samples, outcomes): Stage 1
-        predictions from ridge regression reduction. As defined by
+        predictions from ridge regression reduction. Defined by
         :data:`sgkit.variables.base_prediction`.
 
     - `meta_prediction` (samples, outcomes): Stage 2 predictions from
         the best meta estimator trained on the out-of-sample Stage 1
-        predictions. As defined by :data:`sgkit.variables.meta_prediction`.
+        predictions. Defined by :data:`sgkit.variables.meta_prediction`.
 
     - `loco_prediction` (contigs, samples, outcomes): LOCO predictions
         resulting from Stage 2 predictions ignoring effects for variant
         blocks on held out contigs. This will be absent if the
-        data provided does not contain at least 2 contigs. As defined by
+        data provided does not contain at least 2 contigs. Defined by
         :data:`sgkit.variables.loco_prediction`.
 
     Raises