feat: New primary error type for zero p-values (EBISPOT#42)

* feat: New primary error type for zero p-values

* docs: Update how to run tests

README.md

@@ -181,8 +181,10 @@ To run the tests:
 1. [install poetry](https://python-poetry.org/docs/#installation)
 2. `git clone https://github.com/EBISPOT/gwas-sumstats-tools.git`
 3. `cd gwas-sumstats-tools`
+3. `python3 -m venv env`
+3. `pip install poetry`
 4. `poetry install`
-5. `poetry run pytest`
+5. `poetry run pytest -s`
 
 To make a change:
 branch from master -> PR to master -> poetry version -> git add pyproject.toml -> git commit -> git tag <version> -> git push origin master --tags

gwas_sumstats_tools/schema/data_table.py

@@ -39,6 +39,7 @@ class SumStatsSchema:
             ])
         }
     PVALUE_FIELD_DEFINITIONS = {
+        # 'p_value': Column(float, [
         'p_value': Column(float if platform.system() == "Windows" else Float128, [
             Check.in_range(0, 1,
                            include_min=True,

gwas_sumstats_tools/validate.py

@@ -159,6 +159,8 @@ def _evaluate_errors(self) -> None:
                 self.primary_error_type = 'headers' 
             elif 'Column' in self.errors_table['schema_context']:
                 self.primary_error_type = 'data'
+                if '_p_value_mantissa' in self.errors_table['column']:
+                    self.primary_error_type = 'p_val'                    
 
 
 def validate(filename: Path,

tests/test_validate.py

@@ -210,7 +210,8 @@ def test_zero_pvalue(self, sumstats_file):
         sumstats_file.to_file()
         v = Validator(sumstats_file=sumstats_file.filepath, minimum_rows=4)
         assert v.validate()[0] is False
-        assert v.primary_error_type == "data"
+        # print(v.errors_table)
+        assert v.primary_error_type == "p_val"
         assert v.errors_table.nrows() == 4
         v = Validator(sumstats_file=sumstats_file.filepath, minimum_rows=4,
                       pval_zero=True)
@@ -293,7 +294,7 @@ def test_invalid_pvalue(self, sumstats_file):
         sumstats_file.to_file()
         v = Validator(sumstats_file=sumstats_file.filepath, minimum_rows=4)
         assert v.validate()[0] is False
-        assert v.primary_error_type == "data"
+        assert v.primary_error_type == "p_val"
         print(v.errors_table.look(limit=10))
         assert v.errors_table.nrows() == 6