aboutsummaryrefslogtreecommitdiff
path: root/gn3
diff options
context:
space:
mode:
Diffstat (limited to 'gn3')
-rw-r--r--gn3/db/traits.py69
-rw-r--r--gn3/heatmaps.py67
2 files changed, 70 insertions, 66 deletions
diff --git a/gn3/db/traits.py b/gn3/db/traits.py
index f2673c8..1e29aff 100644
--- a/gn3/db/traits.py
+++ b/gn3/db/traits.py
@@ -1,12 +1,81 @@
"""This class contains functions relating to trait data manipulation"""
import os
+from functools import reduce
from typing import Any, Dict, Union, Sequence
+
from gn3.settings import TMPDIR
from gn3.random import random_string
from gn3.function_helpers import compose
from gn3.db.datasets import retrieve_trait_dataset
+def export_trait_data(
+ trait_data: dict, samplelist: Sequence[str], dtype: str = "val",
+ var_exists: bool = False, n_exists: bool = False):
+ """
+ Export data according to `samplelist`. Mostly used in calculating
+ correlations.
+
+ DESCRIPTION:
+ Migrated from
+ https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/base/webqtlTrait.py#L166-L211
+
+ PARAMETERS
+ trait: (dict)
+ The dictionary of key-value pairs representing a trait
+ samplelist: (list)
+ A list of sample names
+ dtype: (str)
+ ... verify what this is ...
+ var_exists: (bool)
+ A flag indicating existence of variance
+ n_exists: (bool)
+ A flag indicating existence of ndata
+ """
+ def __export_all_types(tdata, sample):
+ sample_data = []
+ if tdata[sample]["value"]:
+ sample_data.append(tdata[sample]["value"])
+ if var_exists:
+ if tdata[sample]["variance"]:
+ sample_data.append(tdata[sample]["variance"])
+ else:
+ sample_data.append(None)
+ if n_exists:
+ if tdata[sample]["ndata"]:
+ sample_data.append(tdata[sample]["ndata"])
+ else:
+ sample_data.append(None)
+ else:
+ if var_exists and n_exists:
+ sample_data += [None, None, None]
+ elif var_exists or n_exists:
+ sample_data += [None, None]
+ else:
+ sample_data.append(None)
+
+ return tuple(sample_data)
+
+ def __exporter(accumulator, sample):
+ # pylint: disable=[R0911]
+ if sample in trait_data["data"]:
+ if dtype == "val":
+ return accumulator + (trait_data["data"][sample]["value"], )
+ if dtype == "var":
+ return accumulator + (trait_data["data"][sample]["variance"], )
+ if dtype == "N":
+ return accumulator + (trait_data["data"][sample]["ndata"], )
+ if dtype == "all":
+ return accumulator + __export_all_types(trait_data["data"], sample)
+ raise KeyError("Type `%s` is incorrect" % dtype)
+ if var_exists and n_exists:
+ return accumulator + (None, None, None)
+ if var_exists or n_exists:
+ return accumulator + (None, None)
+ return accumulator + (None,)
+
+ return reduce(__exporter, samplelist, tuple())
+
def get_trait_csv_sample_data(conn: Any,
trait_name: int, phenotype_id: int):
"""Fetch a trait and return it as a csv string"""
diff --git a/gn3/heatmaps.py b/gn3/heatmaps.py
index 2dd9d07..bf9dfd1 100644
--- a/gn3/heatmaps.py
+++ b/gn3/heatmaps.py
@@ -14,6 +14,7 @@ from plotly.subplots import make_subplots # type: ignore
from gn3.settings import TMPDIR
from gn3.random import random_string
from gn3.computations.slink import slink
+from gn3.db.traits import export_trait_data
from gn3.computations.correlations2 import compute_correlation
from gn3.db.genotypes import (
build_genotype_file, load_genotype_samples)
@@ -26,72 +27,6 @@ from gn3.computations.qtlreaper import (
parse_reaper_main_results,
organise_reaper_main_results)
-def export_trait_data(
- trait_data: dict, samplelist: Sequence[str], dtype: str = "val",
- var_exists: bool = False, n_exists: bool = False):
- """
- Export data according to `samplelist`. Mostly used in calculating
- correlations.
-
- DESCRIPTION:
- Migrated from
- https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/base/webqtlTrait.py#L166-L211
-
- PARAMETERS
- trait: (dict)
- The dictionary of key-value pairs representing a trait
- samplelist: (list)
- A list of sample names
- dtype: (str)
- ... verify what this is ...
- var_exists: (bool)
- A flag indicating existence of variance
- n_exists: (bool)
- A flag indicating existence of ndata
- """
- def __export_all_types(tdata, sample):
- sample_data = []
- if tdata[sample]["value"]:
- sample_data.append(tdata[sample]["value"])
- if var_exists:
- if tdata[sample]["variance"]:
- sample_data.append(tdata[sample]["variance"])
- else:
- sample_data.append(None)
- if n_exists:
- if tdata[sample]["ndata"]:
- sample_data.append(tdata[sample]["ndata"])
- else:
- sample_data.append(None)
- else:
- if var_exists and n_exists:
- sample_data += [None, None, None]
- elif var_exists or n_exists:
- sample_data += [None, None]
- else:
- sample_data.append(None)
-
- return tuple(sample_data)
-
- def __exporter(accumulator, sample):
- # pylint: disable=[R0911]
- if sample in trait_data["data"]:
- if dtype == "val":
- return accumulator + (trait_data["data"][sample]["value"], )
- if dtype == "var":
- return accumulator + (trait_data["data"][sample]["variance"], )
- if dtype == "N":
- return accumulator + (trait_data["data"][sample]["ndata"], )
- if dtype == "all":
- return accumulator + __export_all_types(trait_data["data"], sample)
- raise KeyError("Type `%s` is incorrect" % dtype)
- if var_exists and n_exists:
- return accumulator + (None, None, None)
- if var_exists or n_exists:
- return accumulator + (None, None)
- return accumulator + (None,)
-
- return reduce(__exporter, samplelist, tuple())
def trait_display_name(trait: Dict):
"""