From 157df453cdb84591cb44af9f1d2677cd0b2c0380 Mon Sep 17 00:00:00 2001 From: Frederick Muriuki Muriithi Date: Mon, 18 Oct 2021 12:17:11 +0300 Subject: Move 'export_trait_data' to 'gn3.db.traits' module Issue: https://github.com/genenetwork/gn-gemtext-threads/blob/main/topics/gn1-migration-to-gn2/partial-correlations.gmi * gn3/db/traits.py: Move function `export_trait_data` here * gn3/heatmaps.py: Remove function `export_trait_data` * tests/unit/db/test_traits.py: Move function `export_trait_data` tests here * tests/unit/test_heatmaps.py: Remove function `export_trait_data` here Function `export_trait_data` more closely corresponds to the traits and is used in more than just the `gn3.heatmaps` module. This commit moves the relevant code over to the `gn3.db.traits` module and also moves the tests to the corresponding tests modules. --- gn3/db/traits.py | 69 ++++++++++++++++++++++++++++++++++ gn3/heatmaps.py | 67 +-------------------------------- tests/unit/db/test_traits.py | 89 ++++++++++++++++++++++++++++++++++++++++++++ tests/unit/test_heatmaps.py | 87 ------------------------------------------- 4 files changed, 159 insertions(+), 153 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 adbfbc6..3b94e88 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): """ diff --git a/tests/unit/db/test_traits.py b/tests/unit/db/test_traits.py index 8af8e82..0c4ef78 100644 --- a/tests/unit/db/test_traits.py +++ b/tests/unit/db/test_traits.py @@ -2,6 +2,7 @@ from unittest import mock, TestCase from gn3.db.traits import ( build_trait_name, + export_trait_data, set_haveinfo_field, update_sample_data, retrieve_trait_info, @@ -12,6 +13,38 @@ from gn3.db.traits import ( retrieve_publish_trait_info, retrieve_probeset_trait_info) +samplelist = ["B6cC3-1", "BXD1", "BXD12", "BXD16", "BXD19", "BXD2"] +trait_data = { + "mysqlid": 36688172, + "data": { + "B6cC3-1": {"sample_name": "B6cC3-1", "value": 7.51879, "variance": None, "ndata": None}, + "BXD1": {"sample_name": "BXD1", "value": 7.77141, "variance": None, "ndata": None}, + "BXD12": {"sample_name": "BXD12", "value": 8.39265, "variance": None, "ndata": None}, + "BXD16": {"sample_name": "BXD16", "value": 8.17443, "variance": None, "ndata": None}, + "BXD19": {"sample_name": "BXD19", "value": 8.30401, "variance": None, "ndata": None}, + "BXD2": {"sample_name": "BXD2", "value": 7.80944, "variance": None, "ndata": None}, + "BXD21": {"sample_name": "BXD21", "value": 8.93809, "variance": None, "ndata": None}, + "BXD24": {"sample_name": "BXD24", "value": 7.99415, "variance": None, "ndata": None}, + "BXD27": {"sample_name": "BXD27", "value": 8.12177, "variance": None, "ndata": None}, + "BXD28": {"sample_name": "BXD28", "value": 7.67688, "variance": None, "ndata": None}, + "BXD32": {"sample_name": "BXD32", "value": 7.79062, "variance": None, "ndata": None}, + "BXD39": {"sample_name": "BXD39", "value": 8.27641, "variance": None, "ndata": None}, + "BXD40": {"sample_name": "BXD40", "value": 8.18012, "variance": None, "ndata": None}, + "BXD42": {"sample_name": "BXD42", "value": 7.82433, "variance": None, "ndata": None}, + "BXD6": {"sample_name": "BXD6", "value": 8.09718, "variance": None, "ndata": None}, + "BXH14": {"sample_name": "BXH14", "value": 7.97475, "variance": None, "ndata": None}, + "BXH19": {"sample_name": "BXH19", "value": 7.67223, "variance": None, "ndata": None}, + "BXH2": {"sample_name": "BXH2", "value": 7.93622, "variance": None, "ndata": None}, + "BXH22": {"sample_name": "BXH22", "value": 7.43692, "variance": None, "ndata": None}, + "BXH4": {"sample_name": "BXH4", "value": 7.96336, "variance": None, "ndata": None}, + "BXH6": {"sample_name": "BXH6", "value": 7.75132, "variance": None, "ndata": None}, + "BXH7": {"sample_name": "BXH7", "value": 8.12927, "variance": None, "ndata": None}, + "BXH8": {"sample_name": "BXH8", "value": 6.77338, "variance": None, "ndata": None}, + "BXH9": {"sample_name": "BXH9", "value": 8.03836, "variance": None, "ndata": None}, + "C3H/HeJ": {"sample_name": "C3H/HeJ", "value": 7.42795, "variance": None, "ndata": None}, + "C57BL/6J": {"sample_name": "C57BL/6J", "value": 7.50606, "variance": None, "ndata": None}, + "DBA/2J": {"sample_name": "DBA/2J", "value": 7.72588, "variance": None, "ndata": None}}} + class TestTraitsDBFunctions(TestCase): "Test cases for traits functions" @@ -226,3 +259,59 @@ class TestTraitsDBFunctions(TestCase): with self.subTest(trait_info=trait_info, expected=expected): self.assertEqual( set_confidential_field(trait_type, trait_info), expected) + + def test_export_trait_data_dtype(self): + """ + Test `export_trait_data` with different values for the `dtype` keyword + argument + """ + for dtype, expected in [ + ["val", (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["var", (None, None, None, None, None, None)], + ["N", (None, None, None, None, None, None)], + ["all", (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)]]: + with self.subTest(dtype=dtype): + self.assertEqual( + export_trait_data(trait_data, samplelist, dtype=dtype), + expected) + + def test_export_trait_data_dtype_all_flags(self): + """ + Test `export_trait_data` with different values for the `dtype` keyword + argument and the different flags set up + """ + for dtype, vflag, nflag, expected in [ + ["val", False, False, + (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["val", False, True, + (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["val", True, False, + (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["val", True, True, + (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["var", False, False, (None, None, None, None, None, None)], + ["var", False, True, (None, None, None, None, None, None)], + ["var", True, False, (None, None, None, None, None, None)], + ["var", True, True, (None, None, None, None, None, None)], + ["N", False, False, (None, None, None, None, None, None)], + ["N", False, True, (None, None, None, None, None, None)], + ["N", True, False, (None, None, None, None, None, None)], + ["N", True, True, (None, None, None, None, None, None)], + ["all", False, False, + (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], + ["all", False, True, + (7.51879, None, 7.77141, None, 8.39265, None, 8.17443, None, + 8.30401, None, 7.80944, None)], + ["all", True, False, + (7.51879, None, 7.77141, None, 8.39265, None, 8.17443, None, + 8.30401, None, 7.80944, None)], + ["all", True, True, + (7.51879, None, None, 7.77141, None, None, 8.39265, None, None, + 8.17443, None, None, 8.30401, None, None, 7.80944, None, None)] + ]: + with self.subTest(dtype=dtype, vflag=vflag, nflag=nflag): + self.assertEqual( + export_trait_data( + trait_data, samplelist, dtype=dtype, var_exists=vflag, + n_exists=nflag), + expected) diff --git a/tests/unit/test_heatmaps.py b/tests/unit/test_heatmaps.py index 7b66688..03fd4a6 100644 --- a/tests/unit/test_heatmaps.py +++ b/tests/unit/test_heatmaps.py @@ -4,43 +4,12 @@ from gn3.heatmaps import ( cluster_traits, get_loci_names, get_lrs_from_chr, - export_trait_data, compute_traits_order, retrieve_samples_and_values, process_traits_data_for_heatmap) from tests.unit.sample_test_data import organised_trait_1, organised_trait_2 samplelist = ["B6cC3-1", "BXD1", "BXD12", "BXD16", "BXD19", "BXD2"] -trait_data = { - "mysqlid": 36688172, - "data": { - "B6cC3-1": {"sample_name": "B6cC3-1", "value": 7.51879, "variance": None, "ndata": None}, - "BXD1": {"sample_name": "BXD1", "value": 7.77141, "variance": None, "ndata": None}, - "BXD12": {"sample_name": "BXD12", "value": 8.39265, "variance": None, "ndata": None}, - "BXD16": {"sample_name": "BXD16", "value": 8.17443, "variance": None, "ndata": None}, - "BXD19": {"sample_name": "BXD19", "value": 8.30401, "variance": None, "ndata": None}, - "BXD2": {"sample_name": "BXD2", "value": 7.80944, "variance": None, "ndata": None}, - "BXD21": {"sample_name": "BXD21", "value": 8.93809, "variance": None, "ndata": None}, - "BXD24": {"sample_name": "BXD24", "value": 7.99415, "variance": None, "ndata": None}, - "BXD27": {"sample_name": "BXD27", "value": 8.12177, "variance": None, "ndata": None}, - "BXD28": {"sample_name": "BXD28", "value": 7.67688, "variance": None, "ndata": None}, - "BXD32": {"sample_name": "BXD32", "value": 7.79062, "variance": None, "ndata": None}, - "BXD39": {"sample_name": "BXD39", "value": 8.27641, "variance": None, "ndata": None}, - "BXD40": {"sample_name": "BXD40", "value": 8.18012, "variance": None, "ndata": None}, - "BXD42": {"sample_name": "BXD42", "value": 7.82433, "variance": None, "ndata": None}, - "BXD6": {"sample_name": "BXD6", "value": 8.09718, "variance": None, "ndata": None}, - "BXH14": {"sample_name": "BXH14", "value": 7.97475, "variance": None, "ndata": None}, - "BXH19": {"sample_name": "BXH19", "value": 7.67223, "variance": None, "ndata": None}, - "BXH2": {"sample_name": "BXH2", "value": 7.93622, "variance": None, "ndata": None}, - "BXH22": {"sample_name": "BXH22", "value": 7.43692, "variance": None, "ndata": None}, - "BXH4": {"sample_name": "BXH4", "value": 7.96336, "variance": None, "ndata": None}, - "BXH6": {"sample_name": "BXH6", "value": 7.75132, "variance": None, "ndata": None}, - "BXH7": {"sample_name": "BXH7", "value": 8.12927, "variance": None, "ndata": None}, - "BXH8": {"sample_name": "BXH8", "value": 6.77338, "variance": None, "ndata": None}, - "BXH9": {"sample_name": "BXH9", "value": 8.03836, "variance": None, "ndata": None}, - "C3H/HeJ": {"sample_name": "C3H/HeJ", "value": 7.42795, "variance": None, "ndata": None}, - "C57BL/6J": {"sample_name": "C57BL/6J", "value": 7.50606, "variance": None, "ndata": None}, - "DBA/2J": {"sample_name": "DBA/2J", "value": 7.72588, "variance": None, "ndata": None}}} slinked = ( (((0, 2, 0.16381088984330505), @@ -55,62 +24,6 @@ slinked = ( class TestHeatmap(TestCase): """Class for testing heatmap computation functions""" - def test_export_trait_data_dtype(self): - """ - Test `export_trait_data` with different values for the `dtype` keyword - argument - """ - for dtype, expected in [ - ["val", (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["var", (None, None, None, None, None, None)], - ["N", (None, None, None, None, None, None)], - ["all", (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)]]: - with self.subTest(dtype=dtype): - self.assertEqual( - export_trait_data(trait_data, samplelist, dtype=dtype), - expected) - - def test_export_trait_data_dtype_all_flags(self): - """ - Test `export_trait_data` with different values for the `dtype` keyword - argument and the different flags set up - """ - for dtype, vflag, nflag, expected in [ - ["val", False, False, - (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["val", False, True, - (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["val", True, False, - (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["val", True, True, - (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["var", False, False, (None, None, None, None, None, None)], - ["var", False, True, (None, None, None, None, None, None)], - ["var", True, False, (None, None, None, None, None, None)], - ["var", True, True, (None, None, None, None, None, None)], - ["N", False, False, (None, None, None, None, None, None)], - ["N", False, True, (None, None, None, None, None, None)], - ["N", True, False, (None, None, None, None, None, None)], - ["N", True, True, (None, None, None, None, None, None)], - ["all", False, False, - (7.51879, 7.77141, 8.39265, 8.17443, 8.30401, 7.80944)], - ["all", False, True, - (7.51879, None, 7.77141, None, 8.39265, None, 8.17443, None, - 8.30401, None, 7.80944, None)], - ["all", True, False, - (7.51879, None, 7.77141, None, 8.39265, None, 8.17443, None, - 8.30401, None, 7.80944, None)], - ["all", True, True, - (7.51879, None, None, 7.77141, None, None, 8.39265, None, None, - 8.17443, None, None, 8.30401, None, None, 7.80944, None, None)] - ]: - with self.subTest(dtype=dtype, vflag=vflag, nflag=nflag): - self.assertEqual( - export_trait_data( - trait_data, samplelist, dtype=dtype, var_exists=vflag, - n_exists=nflag), - expected) - def test_cluster_traits(self): """ Test that the clustering is working as expected. -- cgit v1.2.3