Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

update: Fixes to allow compatability with newer numpy & pandas #191

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

update: Fixes to allow compatability with newer numpy & pandas #191

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
604db06
Update MetaXScan to work with newer numpy (and other libraries.) Fix …
pettyalex Mar 10, 2024
d8558e6
Add sqlalchemy as an extra dependency for running tests.
pettyalex Mar 10, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions software/metax/gwas/GWAS.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,7 +100,7 @@ def load_gwas(source, gwas_format, strict=True, separator=None, skip_until_heade
logging.info("Reading input gwas: %s", source)
if separator is None or separator == "ANY_WHITESPACE":
separator = '\s+'
d = pandas.read_table(source, separator)
d = pandas.read_table(source, sep=separator)

logging.info("Processing input gwas")
d = _rename_columns(d, gwas_format)
Expand Down Expand Up @@ -183,7 +183,7 @@ def _enforce_numeric_columns(d):
for column in _numeric_columns:
if column in d:
a = d[column]
if a.dtype == numpy.object:
if a.dtype == object:
a = [str(x) for x in a]
a = [GWASSpecialHandling.sanitize_component(x) for x in a]
d[column] = numpy.array(a, dtype=numpy.float64)
Expand Down
7 changes: 6 additions & 1 deletion software/metax/gwas/GWASSpecialHandling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,12 @@ def _ogz(p):
s[comp].append(c)

for c in header_comps:
s[c] = numpy.array(pandas.to_numeric(s[c], errors='ignore'))
try:
s[c] = numpy.array(pandas.to_numeric(s[c], errors='raise'))
except Exception as e:
# logging.error("Error converting array to_numeric: ", e)
s[c] = s[c] # This is the behavior "ignore" was doing


return s

Expand Down
8 changes: 4 additions & 4 deletions software/metax/gwas/Utilities.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,11 +89,11 @@ def gwas_from_data(data, extra_columns=None):
else:
rsid, chromosome, position, non_effect_allele, effect_allele, zscore = [], [], [], [], [], []

g = pandas.DataFrame({Constants.SNP:numpy.array(rsid, dtype=numpy.str),
Constants.CHROMOSOME:numpy.array(chromosome, dtype=numpy.str),
g = pandas.DataFrame({Constants.SNP:numpy.array(rsid, dtype=str),
Constants.CHROMOSOME:numpy.array(chromosome, dtype=str),
Constants.POSITION:numpy.array(position),
Constants.EFFECT_ALLELE:numpy.array(effect_allele, dtype=numpy.str),
Constants.NON_EFFECT_ALLELE:numpy.array(non_effect_allele, dtype=numpy.str),
Constants.EFFECT_ALLELE:numpy.array(effect_allele, dtype=str),
Constants.NON_EFFECT_ALLELE:numpy.array(non_effect_allele, dtype=str),
Constants.ZSCORE:numpy.array(zscore)})
if len(data) and extra_columns:
for k,i in extra_columns:
Expand Down
2 changes: 1 addition & 1 deletion software/metax/metaxcan/Utilities.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,7 +115,7 @@ def provide_calculation(self, gene):
while True:
w = self._get_weights(gene)
gwas = self._get_gwas(list(w.keys()))
type = [numpy.str, numpy.float64, numpy.float64, numpy.float64]
type = [str, numpy.float64, numpy.float64, numpy.float64]
columns = [Constants.SNP, WDBQF.K_WEIGHT, Constants.ZSCORE, Constants.BETA]
d = {x: v for x, v in w.items() if x in gwas}

Expand Down
3 changes: 2 additions & 1 deletion software/setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,8 +29,9 @@ def read(fname):
'MulTiXcan.py',
'SMulTiXcan.py'],
description=["TBD"],
install_requires=['scipy>=1.2.2,<1.3', 'numpy>=1.14.2', 'pandas>=0.22.0', 'patsy>=0.5.0',
install_requires=['scipy>=1.2.2', 'numpy>=1.14.2', 'pandas>=0.22.0', 'patsy>=0.5.0',
'statsmodels>=0.10.0', 'h5py>=2.7.1', 'h5py-cache>=1.0', 'bgen_reader>=3.0.3', 'cyvcf2>=0.8.0'],
extras_require={"test": ["sqlalchemy"]},
long_description=read('Readme.md'),
keywords=['TBD'],
test_suite='tests',
Expand Down
8 changes: 4 additions & 4 deletions software/tests/scz2_sample.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,10 +1,10 @@
import pandas
import numpy

expected_snp = pandas.Series(["rs940550", "rs6650104", "rs6594028", "rs9701055", "rs7417504", "rs12082473", "rs3094315", "rs3131971", "rs61770173", ], dtype=numpy.str)
expected_effect = pandas.Series(["C", "T", "T", "A", "T", "A", "A", "T", "A"], dtype=numpy.str)
expected_non_effect = pandas.Series(["G", "C", "C", "T", "C", "G", "G", "C", "C"], dtype=numpy.str)
expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr22", "chr22"], dtype=numpy.str)
expected_snp = pandas.Series(["rs940550", "rs6650104", "rs6594028", "rs9701055", "rs7417504", "rs12082473", "rs3094315", "rs3131971", "rs61770173", ], dtype=str)
expected_effect = pandas.Series(["C", "T", "T", "A", "T", "A", "A", "T", "A"], dtype=str)
expected_non_effect = pandas.Series(["G", "C", "C", "T", "C", "G", "G", "C", "C"], dtype=str)
expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr22", "chr22"], dtype=str)
expected_position = pandas.Series([729679, 731718, 734349, 736289, 751756, 752566, 752721, 752894, 753405])
expected_se = pandas.Series( [0.0173, 0.0198, 0.02, 0.0193, 0.0164, 0.0149, 0.0146, 0.015, 0.0159], dtype=numpy.float32)
expected_p = pandas.Series( [0.2083, 0.3298, 0.3055, 0.5132, 0.8431, 0.7870, 0.8229, 0.5065, 0.8181], dtype=numpy.float32)
Expand Down
16 changes: 8 additions & 8 deletions software/tests/test_gwas.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,19 +48,19 @@ def assert_gwas_zscore_pb(unit_test, gwas):
numpy.testing.assert_allclose(gwas[PVALUE], scz2_sample.expected_p, rtol=0.001)

def assert_gwas_extracted_from_data_3(unit_test, gwas):
expected_snp = pandas.Series(["rs3", "rs6", "rs7"], dtype=numpy.str)
expected_snp = pandas.Series(["rs3", "rs6", "rs7"], dtype=str)
numpy.testing.assert_array_equal(gwas[SNP], expected_snp)

expected_effect = pandas.Series(["G", "G", "T"], dtype=numpy.str)
expected_effect = pandas.Series(["G", "G", "T"], dtype=str)
numpy.testing.assert_array_equal(gwas[EFFECT_ALLELE], expected_effect)

expected_non_effect = pandas.Series(["A", "A", "C"], dtype=numpy.str)
expected_non_effect = pandas.Series(["A", "A", "C"], dtype=str)
numpy.testing.assert_array_equal(gwas[NON_EFFECT_ALLELE], expected_non_effect)

expected_zscore = pandas.Series([1.3, 2.9, 4.35], dtype=numpy.float32)
numpy.testing.assert_allclose(gwas[ZSCORE], expected_zscore, rtol=0.001)

expected_chromosome = pandas.Series(["chr1", "chr1", "chr1"], dtype=numpy.str)
expected_chromosome = pandas.Series(["chr1", "chr1", "chr1"], dtype=str)
numpy.testing.assert_array_equal(gwas[CHROMOSOME], expected_chromosome)

def _add_basic_to_format(format):
Expand Down Expand Up @@ -169,10 +169,10 @@ def test_gwas_from_source(self):

gwas = GWAS.load_gwas("tests/_td/GWAS/scz2/scz2.gwas.results.txt.gz", gwas_format, snps={"rs940550", "rs6650104", "rs61770173"})

numpy.testing.assert_array_equal(gwas[SNP], pandas.Series(["rs940550", "rs6650104", "rs61770173", ], dtype=numpy.str))
numpy.testing.assert_array_equal(gwas[EFFECT_ALLELE], pandas.Series(["C", "T", "A"], dtype=numpy.str))
numpy.testing.assert_array_equal(gwas[NON_EFFECT_ALLELE], pandas.Series(["G", "C", "C"], dtype=numpy.str))
numpy.testing.assert_array_equal(gwas[CHROMOSOME], pandas.Series(["chr1", "chr1", "chr22"], dtype=numpy.str))
numpy.testing.assert_array_equal(gwas[SNP], pandas.Series(["rs940550", "rs6650104", "rs61770173", ], dtype=str))
numpy.testing.assert_array_equal(gwas[EFFECT_ALLELE], pandas.Series(["C", "T", "A"], dtype=str))
numpy.testing.assert_array_equal(gwas[NON_EFFECT_ALLELE], pandas.Series(["G", "C", "C"], dtype=str))
numpy.testing.assert_array_equal(gwas[CHROMOSOME], pandas.Series(["chr1", "chr1", "chr22"], dtype=str))
numpy.testing.assert_allclose(gwas[ZSCORE], pandas.Series([-1.254557, 0.974874, -0.232505],dtype=numpy.float32), rtol=0.001)
numpy.testing.assert_allclose(gwas[BETA], pandas.Series([-0.0217038334437866, 0.0193025022544974, -0.00369682484428976], dtype=numpy.float32), rtol=0.001)
numpy.testing.assert_allclose(gwas[SE], pandas.Series([0.0173, 0.0198, 0.0159], dtype=numpy.float32), rtol=0.001)
Expand Down
16 changes: 8 additions & 8 deletions software/tests/test_gwas_utilities.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,38 +15,38 @@
from . import SampleData

def assert_gwas_1(unit_test, gwas):
expected_snp = pandas.Series(["rs1666", "rs1", "rs2", "rs3", "rs4", "rs6", "rs7", "rs7666", "rs8", "rs9"], dtype=numpy.str)
expected_snp = pandas.Series(["rs1666", "rs1", "rs2", "rs3", "rs4", "rs6", "rs7", "rs7666", "rs8", "rs9"], dtype=str)
numpy.testing.assert_array_equal(gwas[SNP], expected_snp)

expected_effect = pandas.Series(["A", "C", "C", "G", "A", "G", "T", "A", "A", "A"], dtype=numpy.str)
expected_effect = pandas.Series(["A", "C", "C", "G", "A", "G", "T", "A", "A", "A"], dtype=str)
numpy.testing.assert_array_equal(gwas[EFFECT_ALLELE], expected_effect)

expected_non_effect = pandas.Series(["G", "T", "T", "A", "G", "A", "C", "G", "G", "G"], dtype=numpy.str)
expected_non_effect = pandas.Series(["G", "T", "T", "A", "G", "A", "C", "G", "G", "G"], dtype=str)
numpy.testing.assert_array_equal(gwas[NON_EFFECT_ALLELE], expected_non_effect)

expected_zscore = pandas.Series([0.3, -0.2, 0.5, 1.3, -0.3, 2.9, 4.35, 1.3, 0.09, 0.09], dtype=numpy.float32)
numpy.testing.assert_allclose(gwas[ZSCORE], expected_zscore, rtol=0.001)

expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1"], dtype=numpy.str)
expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1"], dtype=str)
numpy.testing.assert_array_equal(gwas[CHROMOSOME], expected_chromosome)

expected_position = pandas.Series([0, 1, 5, 20, 30, 42, 43, 45, 50, 70])
numpy.testing.assert_array_equal(gwas[POSITION], expected_position)

def assert_gwas_2(unit_test, gwas):
expected_snp = pandas.Series(["rsC", "rs1666", "rs1", "rs2", "rs4", "rsB", "rsA", "rs7666", "rs8", "rs9"], dtype=numpy.str)
expected_snp = pandas.Series(["rsC", "rs1666", "rs1", "rs2", "rs4", "rsB", "rsA", "rs7666", "rs8", "rs9"], dtype=str)
numpy.testing.assert_array_equal(gwas[SNP], expected_snp)

expected_effect = pandas.Series(["T", "A", "C", "C", "A", "G", "G", "A", "A", "A"], dtype=numpy.str)
expected_effect = pandas.Series(["T", "A", "C", "C", "A", "G", "G", "A", "A", "A"], dtype=str)
numpy.testing.assert_array_equal(gwas[EFFECT_ALLELE], expected_effect)

expected_non_effect = pandas.Series(["C", "G", "T", "T", "G", "A", "A", "G", "G", "G"], dtype=numpy.str)
expected_non_effect = pandas.Series(["C", "G", "T", "T", "G", "A", "A", "G", "G", "G"], dtype=str)
numpy.testing.assert_array_equal(gwas[NON_EFFECT_ALLELE], expected_non_effect)

expected_zscore = pandas.Series([4.35, 0.3, -0.2, 1.3, -0.3, 2.9, 1.3, 1.3, 0.09, 0.09], dtype=numpy.float32)
numpy.testing.assert_allclose(gwas[ZSCORE], expected_zscore, rtol=0.001)

expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1"], dtype=numpy.str)
expected_chromosome = pandas.Series(["chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1", "chr1"], dtype=str)
numpy.testing.assert_array_equal(gwas[CHROMOSOME], expected_chromosome)

expected_position = pandas.Series([None, None, None, None, None, None, None, None, None, None])
Expand Down