Merge pull request #386 from monarch-initiative/release-0.9.1

Make release `0.9.1`

Author: ielis

docs/conf.py

@@ -63,7 +63,7 @@
 # The short X.Y version.
 version = u'0.9'
 # The full version, including alpha/beta/rc tags.
-release = u'0.9.0'
+release = u'0.9.1'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.

docs/tutorial.rst

@@ -216,7 +216,7 @@ The diagram also highlights the protein features (domains, repeats, etc.).
 Summarize all variant alleles
 -----------------------------
 
-We can prepare a table of all variant alleles that occurr in the cohort.
+We can prepare a table of all variant alleles that occur in the cohort.
 
 Each table row corresponds to a single allele and lists the variant key,
 the predicted effect on the transcript (*cDNA*) and protein of interest,
@@ -240,11 +240,25 @@ with one or more variant alleles (*Count*):
 Partition the cohort by genotype and phenotype
 ==============================================
 
-To test for genotype-phenotype associations, we need to divide the cohort into classes.
-In GPSEA, we always assign a cohort member into a genotype class,
-where each individual is assigned into a single class and the classes do not overlap.
-The phenotype is then used to either assign an individual into a class,
-or to calculate a numeric score or survival.
+Testing for a genotype-phenotype association uses genotype and phenotype as variables.
+In GPSEA, the variable value for an individual is computed
+either by a :class:`~gpsea.analysis.clf.Classifier`
+or by a :class:`~gpsea.analysis.pscore.PhenotypeScorer`.
+A `Classifier` assigns the individual into a class,
+whereas a `PhenotypeScorer` computes a continuous score.
+The classifiers and scorers are applied on all individuals of the cohort
+and the resulting variable distributions are then assessed by a statistical test.
+
+In GPSEA, genotype is always treated as a class
+and a genotype `Classifier` is a prerequisite for each analysis.
+However, there is much more flexibility on the phenotype part,
+where either a `Classifier` or a `PhenotypeScorer` can be used to compute the values,
+depending on the analysis goals.
+
+In this tutorial section, we first configure a `Classifier` for assigning
+the individuals into a genotype class,
+and we follow with generating classifiers for testing the presence or exclusion
+of HPO terms in the individuals.
 
 
 Partition by genotype

docs/user-guide/analyses/partitioning/phenotype/hpo_predicate.rst

@@ -71,7 +71,7 @@ function cuts down the tedium.
 Example
 -------
 
-For a phenopacket collection (e.g. 156 patients with mutations in *WWOX* gene included in Phenopacket Store version `0.1.18`)
+For a phenopacket collection (e.g. 156 patients with mutations in *TBX5* gene included in Phenopacket Store version `0.1.18`)
 
 >>> from ppktstore.registry import configure_phenopacket_registry
 >>> registry = configure_phenopacket_registry()

src/gpsea/__init__.py

@@ -2,7 +2,7 @@
 GPSEA is a library for finding genotype-phenotype associations.
 """
 
-__version__ = "0.9.0"
+__version__ = "0.9.1"
 
 _overwrite = False
 """

src/gpsea/view/_protein_visualizer.py

@@ -510,11 +510,7 @@ def draw_axes(ax, x_ticks, x_ticks_relative, y_ticks, max_marker_count,
             "# Variants", y_axis_x - 0.05, (y_axis_min_y + y_axis_max_y) / 2,
             fontsize=font_size, ha='center', va='center', rotation=90,
         )
-        draw_string(  # x axis label
-            ax,
-            "# Codons", (x_axis_min_x + x_axis_max_x) / 2, x_axis_y - 0.05,
-            fontsize=font_size, ha='center', va='center', rotation=0,
-                          )
+        # Note that we do not label the X-axis, the meaning will be obvious to users (amino acid residues)
         # draw y ticks
         draw_line(  # 0 tick
             ax,