You're Invited:Meet the Socket Team at RSAC and BSidesSF 2026, March 23–26.RSVP →
Socket
Book a DemoSign in
Socket
Book a DemoSign in

gumpy

Package Overview
Dependencies
Maintainers
2
Versions
34
Alerts
File Explorer

Advanced tools

License
Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

gumpy - pypi Package Compare versions

Comparing version
1.2.7
 to
1.3.0
+1
-1
gumpy.egg-info/PKG-INFO
Metadata-Version: 2.1
Name: gumpy
Version: 1.2.7
Version: 1.3.0
Summary: Genetics with Numpy

@@ -5,0 +5,0 @@ Home-page: https://github.com/oxfordmmm/gumpy

+72
-18
gumpy/variantfile.py

@@ -40,3 +40,5 @@ """

def __init__(self, record: pysam.libcbcf.VariantRecord, sample: str | int):
def __init__(
self, record: pysam.libcbcf.VariantRecord, sample: str | int, min_dp: int | None
):
"""Constructor for the VCFRecord object.

@@ -50,2 +52,3 @@

where there is more than 1 sample per record
min_dp (int | None): Minimum depth to consider a call.
"""

@@ -101,9 +104,19 @@

if key == "GT":
call1, call2 = item
self.call1 = call1 if call1 is not None else -1
self.call2 = self.call1 if call2 is None else call2
if len(item) == 2:
# Ploidy 2 is expected here
call1, call2 = item
self.call1 = call1 if call1 is not None else -1
self.call2 = self.call1 if call2 is None else call2
else:
# GVCF edge case with ploidy 1
call1 = item[0]
self.call1 = call1 if call1 is not None else -1
self.call2 = self.call1
self.is_reference = (
True if call1 == 0 and (call1 == call2 or call2 == -1) else False
True
if self.call1 == 0
and (self.call1 == self.call2 or self.call2 == -1)
else False
)
self.is_heterozygous = True if call1 != call2 else False
self.is_heterozygous = True if self.call1 != self.call2 else False
self.is_null = True if set([self.call1, self.call2]) == {-1} else False

@@ -125,2 +138,27 @@ self.is_alt = (

if min_dp is not None:
allelic_depth_tag = "COV" if "COV" in self.values.keys() else "AD"
# Ensure we have a COV tag for downstream analysis
self.values["COV"] = self.values[allelic_depth_tag]
print(self.values[allelic_depth_tag])
if self.values[allelic_depth_tag] != (None,):
# If the depth given is below the threshold,
# this row is a null call's row from the GVCF
# So treat it as such.
if len(self.values[allelic_depth_tag]) == 1:
if self.values[allelic_depth_tag][0] < min_dp:
# Odd case where GVCF only gives the depth of the alt
self.is_null = True
self.is_heterozygous = False
self.is_alt = False
self.is_reference = False
elif (
self.values[allelic_depth_tag][self.call1] < min_dp
or self.values[allelic_depth_tag][self.call2] < min_dp
):
self.is_null = True
self.is_heterozygous = False
self.is_alt = False
self.is_reference = False
def __repr__(self) -> str:

@@ -203,2 +241,3 @@ """Pretty print the record

minor_population_indices: Collection[int] = [],
min_dp: int | None = None,
):

@@ -221,2 +260,3 @@ """

within which to look for minor populations
min_dp (int, optional): Minimum depth to consider a call. Default is None
"""

@@ -293,3 +333,3 @@

for sample in record.samples.keys():
self.records.append(VCFRecord(record, sample))
self.records.append(VCFRecord(record, sample, min_dp))

@@ -310,3 +350,6 @@ # Find calls will ensure that no calls have same position

or "AD" in self.format_fields_metadata.keys()
), "'COV' and 'AD' not in VCF format fields. No minor populations can be found!"
), (
"'COV' and 'AD' not in VCF format fields. "
"No minor populations can be found!"
)
self._find_minor_populations()

@@ -390,6 +433,2 @@ else:

allelic_depth_tag = (
"COV" if "COV" in self.format_fields_metadata.keys() else "AD"
)
for idx, type_ in self.calls.keys():

@@ -402,2 +441,10 @@ # Check if we've delt with this vcf already

# Pull out depth tag from the specific row's format fields
# as the file metadata isn't a guarantee of the actual fields of this row
allelic_depth_tag = (
"COV"
if self.calls[(idx, type_)]["original_vcf_row"].get("COV", None)
else "AD"
)
# Checking for het calls

@@ -418,6 +465,10 @@ if self.calls[(idx, type_)]["call"] == "z":

# Get all of the calls
calls = [self.calls[(idx, type_)]["original_vcf_row"]["REF"]] + list(
self.calls[(idx, type_)]["original_vcf_row"]["ALTS"]
)
if self.calls[(idx, type_)]["original_vcf_row"]["ALTS"] is None:
# Case arrises from gvcf ref calls not giving any alts
calls = [self.calls[(idx, type_)]["original_vcf_row"]["REF"]]
else:
# Get all of the calls
calls = [self.calls[(idx, type_)]["original_vcf_row"]["REF"]] + list(
self.calls[(idx, type_)]["original_vcf_row"]["ALTS"]
)

@@ -430,4 +481,7 @@ # Break down the calls as appropriate

# total_depth = self.calls[(idx, type_)]['original_vcf_row']['DP']
total_depth = sum(dps)
# GVCF null calls get None for depth, so catch (and skip) this
if dps == [None]:
continue
else:
total_depth = sum(dps)

@@ -434,0 +488,0 @@ # idx here refers to the position of this call, NOT this vcf row, so adjust

+1
-1
PKG-INFO
Metadata-Version: 2.1
Name: gumpy
Version: 1.2.7
Version: 1.3.0
Summary: Genetics with Numpy

@@ -5,0 +5,0 @@ Home-page: https://github.com/oxfordmmm/gumpy

+2
-1
VERSION

@@ -1,1 +0,2 @@

1.2.7
1.3.0