Fix versions and packages dependencies

Author: AndreRicoPSU

clarite/internal/calculations.py

@@ -1,9 +1,9 @@
+import re
+
 import numpy as np
 from scipy import stats
 from scipy.optimize import brentq as root
 
-import re
-
 from clarite.modules.analyze.utils import statsmodels_var_regex
 
 
@@ -30,7 +30,7 @@ def regTermTest(full_model, restricted_model, ddf, X_names, var_name):
 
 def _pFsum(x, df, a, ddf):
     tr = a.mean()
-    tr2 = (a ** 2).mean() / tr ** 2
+    tr2 = (a**2).mean() / tr**2
     scale = tr * tr2
     ndf = len(a) / tr2
 
@@ -63,7 +63,7 @@ def kprime0(z):
         return (lam / (1 - 2 * z * lam)).sum()
 
     def kpprime0(z):
-        return 2 * (lam ** 2 / (1 - 2 * z * lam) ** 2).sum()
+        return 2 * (lam**2 / (1 - 2 * z * lam) ** 2).sum()
 
     if any(lam < 0):
         lam_min = (1 / (2 * lam[lam < 0])).max() * 0.99999

clarite/modules/analyze/regression/glm_regression.py

@@ -127,7 +127,9 @@ def __init__(
             counts = self.data[self.outcome_variable].value_counts().to_dict()
 
             categories = self.data[self.outcome_variable].cat.categories
-            categories = sorted(categories, reverse=True)  # Sort reverse to keep control as 0 and case as 1
+            # Sort reverse to keep control as 0 and case as 1
+            if categories[0] == "Case" and categories[1] == "Control":
+                categories = sorted(categories, reverse=True)
             codes, categories = zip(*enumerate(categories))
             self.data[self.outcome_variable].replace(categories, codes, inplace=True)
             self.description += (

clarite/modules/analyze/utils.py

@@ -1,5 +1,5 @@
 import re
-from typing import Optional, List, Union
+from typing import List, Optional, Union
 
 import numpy as np
 import pandas as pd
@@ -148,4 +148,4 @@ def statsmodels_var_regex(variable_name) -> str:
     1) Categoricals may have the value encoded at the end (for example, Male[T.1.0])
     2) Symbols in the variable name must be accounted for: (, ), -, +, etc
     """
-    return fr"^{re.escape(variable_name)}(\[T\..*\])?$"
+    return rf"^{re.escape(variable_name)}(\[T\..*\])?$"

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "clarite"
-version = "2.2.0"
+version = "2.2.2"
 description = "CLeaning to Analysis: Reproducibility-based Interface for Traits and Exposures"
 authors = ["Hall Lab <mah546@psu.edu>"]
 license = "BSD-3-Clause"
@@ -16,12 +16,12 @@ classifiers=[
     ]
 
 [tool.poetry.dependencies]
-python = ">=3.7.1,<3.10"
+python = ">=3.7.1,<3.8"
 click = ">7"
-pandas = ">1.1"
-scipy = ">1.5"
+pandas = "^1.2"
+scipy = "^1.6"
 seaborn = ">0.9"
-statsmodels = ">0.10"
+statsmodels = "^0.12"
 matplotlib = "^3.4.2"
 numpy = "^1.21.0"
 rpy2 = "^3.4.5"
@@ -31,9 +31,9 @@ numpydoc = {version = "^1.1.0", optional = true}
 sphinx_rtd_theme = {version = "^0.5.0", optional = true}
 sphinx-copybutton = {version = "^0.3.0", optional = true}
 ipython = {version = "^7.18.1", optional = true}
-sphinx-click = {version = "^2.6", optional = true}
+sphinx-click = {version = "^4", optional = true}
 importlib-metadata = {version = "^2.0", python = "<3.8"}
-pandas-genomics = ">=0.10.1"
+pandas-genomics = "^0.11"
 
 [tool.poetry.dev-dependencies]
 codecov = "^2.1.11"
@@ -42,7 +42,7 @@ flake8 = "^3.9.2"
 pytest = ">=4.6"
 pytest-cov = "^2.12.1"
 pytest-xdist = "^2.3.0"
-black = "^21.6b0"
+black = ">=22"
 
 [tool.poetry.extras]
 docs = ["sphinx", "numpydoc", "sphinx_rtd_theme", "sphinx-copybutton", "ipython", "sphinx-click"]

tests/analyze/test_association_study.py

@@ -987,15 +987,34 @@ def test_edge_encondig_logistic_regression():
     from pandas_genomics import sim
 
     # Additive Main Effect for SNP1 without interaction
-    train_add_main_effect45 = sim.BAMS.from_model(eff1=sim.SNPEffectEncodings.ADDITIVE, eff2=sim.SNPEffectEncodings.ADDITIVE,
-                                                  penetrance_base=0.45, main1=1, main2=0, interaction=0, random_seed=2021)
-    test_add_main_effect45 = sim.BAMS.from_model(eff1=sim.SNPEffectEncodings.ADDITIVE,
-                                                 eff2=sim.SNPEffectEncodings.ADDITIVE, penetrance_base=0.45, main1=1, main2=0, interaction=0)
-    train_add_me_pb045 = train_add_main_effect45.generate_case_control(n_cases=5000, n_controls=5000)
-    test_add_me_pb045 = test_add_main_effect45.generate_case_control(n_cases=5000, n_controls=5000)
-    edge_weights_add_me_pb045 = train_add_me_pb045.genomics.calculate_edge_encoding_values(data=train_add_me_pb045["Outcome"], outcome_variable="Outcome")
-
-    edge_results_add_me_pb045 = clarite.analyze.association_study(
-        data=test_add_me_pb045, outcomes="Outcome", encoding="edge", edge_encoding_info=edge_weights_add_me_pb045)
-
-    assert 2 == 2
+    train = sim.BAMS.from_model(
+        eff1=sim.SNPEffectEncodings.ADDITIVE,
+        eff2=sim.SNPEffectEncodings.ADDITIVE,
+        penetrance_base=0.45,
+        main1=1,
+        main2=0,
+        interaction=0,
+        random_seed=2021,
+    )
+    test = sim.BAMS.from_model(
+        eff1=sim.SNPEffectEncodings.ADDITIVE,
+        eff2=sim.SNPEffectEncodings.ADDITIVE,
+        penetrance_base=0.45,
+        main1=1,
+        main2=0,
+        interaction=0,
+    )
+    train_add = train.generate_case_control(n_cases=5000, n_controls=5000)
+    test_add = test.generate_case_control(n_cases=5000, n_controls=5000)
+    edge_weights = train_add.genomics.calculate_edge_encoding_values(
+        data=train_add["Outcome"], outcome_variable="Outcome"
+    )
+
+    edge_results = clarite.analyze.association_study(
+        data=test_add,
+        outcomes="Outcome",
+        encoding="edge",
+        edge_encoding_info=edge_weights,
+    )
+
+    assert edge_results["Variable_type"][0] == "continuous"