def test_get_genes(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
self.assertEqual(len(c.get_genes()), 3, 'Number of genes should be 3')
self.assertEqual(len(c.genes), 3, 'Number of genes should be 3')
self.assertEqual(c.genes, EXAMPLE_MEMBERS[:-1],
'Given and recieved genes do not match')
def test_go_enrichment(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
df = c.go_enrichment()
self.assertNotEqual(df.shape[0], 0, 'Results should not be empty')
self.assertEqual(df.iloc[0, 6], 'GO:0006649',
'Results to not match expected')
self.assertEqual(df.iloc[0, 5], 0.0002912762741932548,
'Results to not match expected')
def test_get_ts(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
c.get_genes()
stuff = c.get_gene_tissue_specificities()
self.assertEqual(
len(stuff[0]), len(c.get_genes()),
'Length of genes and tissue specific information should match')
self.assertEqual(
len(stuff[1]), len(c.get_genes()),
'Length of genes and cell-type group information should match')
self.assertEqual(
len(stuff[2]), len(c.get_genes()),
'Length of genes and cell-type specific information should match')
def test_reset(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
c.get_genes()
self.assertEqual(len(c.genes), 3, 'Number of genes should be 3')
c.reset()
self.assertEqual(c.genes, None, 'Genes should be set to None')
def test_disease_associated_genes(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
c.get_genes()
stuff = c.get_diseases_associated_with_genes()
self.assertEqual(
len(stuff), len(c.get_genes()),
'Length of genes and tissue specific information should match')
def test_blank(self):
c = BOCC()
self.assertEqual(c.name, None,
'Name should be None when first initialized')
self.assertEqual(len(c.members), 0,
'Members should be initialized as an empty list')
self.assertEqual(len(c.members), len(c.types),
'length of members and types should always be equal')
def test_get_summary_stats(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
df = c.get_summary_stats('Data/mygene2_gene_hpo_family.tsv')
self.assertEqual(df.iloc[0, :]['cluster_size'], 4,
'Cluster size does not match expectation')
self.assertEqual(df.iloc[0, :]['gene_ratio'], 0.75,
'gene ratio not as expected')
self.assertEqual(df.iloc[0, :]['HPO_ratio'], 0.25,
'HPO ratio not as expected')
self.assertEqual(df.iloc[0, :]['num_sig_go_enrichment_terms'], 0,
'wrong number of expected significant terms')
self.assertEqual(df.iloc[0, :]['go_sig_threshold'], 0.05,
'wrong threshold for significance')
self.assertEqual(df.iloc[0, :]['max_norm_cell_type_specificity'],
1 / 3, 'wrong max norm cell type specificity')
self.assertEqual(df.iloc[0, :]['max_norm_disease_specificity'], 3 / 3,
'wrong max norm cell type specificity')
def test_get_disease_counts(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
c.get_genes()
stuff = c.get_disease_counts()
self.assertEqual(
len(stuff), 6,
'Number of returned associations does not match expectation')
self.assertEqual(
stuff['Retinitis pigmentosa'], 3,
'All gene in the community (3) should be associated with RP')
def test_summarize_disease_associations(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
stuff = c.summarize_disease_associations()
self.assertEqual(
stuff[1], len(c.get_genes()),
'The max association count should not be larger than the number of genes in the com'
)
from BOCC import BOCC
import matplotlib.pyplot as plt
from matplotlib.lines import Line2D
import numpy as np
import pandas as pd
com_files = ['cesna_coms_june_22_2021.txt', 'greedy_coms_june_22_2021.txt', 'infomap_coms_june_22_2021.txt',
'walktrap_coms_june_22_2021_filtered.txt']
c = BOCC.load_clusters(com_files[0])
t = c[0].mygene2_stats('Data/mygene2_gene_hpo_family.tsv')
fam = t[-1]
# x: total pairs
# y: num rediscovered
# color: aglo
total_num_pairs_per_fam = []
num_rediscovered_per_fam = []
algo = []
families = []
com_id = []
for file in com_files:
coms = BOCC.load_clusters(file)
for c in coms:
t = c.mygene2_stats('Data/mygene2_gene_hpo_family.tsv')
fam = t[-1]
for key in fam.keys():
x = fam[key]
total_num_pairs_per_fam.append(len(x['pairs']) + len(x['not_pairs']))
num_rediscovered_per_fam.append(len(x['pairs']))
algo.append(file)
def test_add_members(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
self.assertEqual(len(c.members), len(EXAMPLE_MEMBERS),
'Number of members should be 4')
self.assertEqual(len(c.types), len(EXAMPLE_TYPES),
'Number of types should be 4')
# add more to the members
c.add_members(['gene1', 'HP:000007'], ['gene', 'hpo'])
self.assertEqual(len(c.members), 6, 'Number of members should be 6')
self.assertEqual(len(c.types), 6, 'Number of types should be 6')
# add more but without types listed
c.add_members(['gene2', 'HP:000008'])
self.assertEqual(len(c.members), 8, 'Number of members should be 8')
self.assertEqual(len(c.types), 8, 'Number of types should be 8')
del c
for line in open(SMALL_TEST_COMS, 'r'):
c = BOCC()
row = line.strip().split('\t')
c.add_members(row[1:])
self.assertEqual(
len(c.members),
len(row) - 1,
'Number of members in BOCC obj doesn\'t match expectation')
def test_summarize_cell_type_specificity(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
stuff = c.summarize_cell_type_specificity()
self.assertEqual(stuff[1], 1,
'The max association count should not be 1')
def test_get_cell_type_counts(self):
c = BOCC()
c.add_members(EXAMPLE_MEMBERS, EXAMPLE_TYPES)
stuff = c.get_cell_type_counts()
self.assertEqual(stuff['Cone photoreceptor cells'], 1,
'There should be 1 Cone cell occurrence')