def my_gene_sets():
gene_sets = GeneSetCollection([
GeneSet('GeneSet1',
'First gene set', ['a', 'b', 'd'],
source='TestSource',
collection='TestCollection',
description='The first test GeneSet.'),
GeneSet('GeneSet2',
'Second gene set', ['a', 'c', 'd'],
source='TestSource',
collection='TestCollection',
description='The second test GeneSet.'),
])
return gene_sets
def my_gene_sets(my_gene_set):
gene_sets = []
for i in range(3):
gs = GeneSet(my_gene_set.id + str(i + 1), my_gene_set.name,
my_gene_set.genes)
gene_sets.append(gs)
return gene_sets
def my_gene_set(my_genes):
gene_set = GeneSet('TestID',
'TestName',
my_genes,
source='TestSource',
collection='TestCollection',
description='Test GeneSet.')
return gene_set
def my_rank_based_result(my_matrix, my_v):
indices = np.uint16(np.nonzero(my_v)[0])
ind_genes = [my_matrix.genes[i] for i in indices]
gs_genes = list(ind_genes)
gene_set = GeneSet(genes=gs_genes, id='Random1', name='Random gene Set 1')
N = my_v.size
X = 1
L = N
## stat, n_star, pval = xlmhg_test(my_v, X, L)
res = get_xlmhg_test_result(N, indices, X, L)
result = RankBasedGSEResult(gene_set, N, indices, ind_genes, X, L,
res.stat, res.cutoff, res.pval)
return result
def my_uninteresting_gene_set(my_ranked_genes):
"""Select the last five ranked genes"""
genes = my_ranked_genes[-5:]
gene_set = GeneSet('BoringID', 'boring gene set', genes)
return gene_set
def my_gene_set(my_ranked_genes, my_v):
"""Select the genes corresponding to the 1's in ``my_v``."""
genes = [my_ranked_genes[i] for i in np.nonzero(my_v)[0]]
gene_set = GeneSet('TestID', 'TestName', genes)
return gene_set
def get_gene_sets(self, min_genes=None, max_genes=None):
"""Return the set of annotated genes for each GO term.
Parameters
----------
min_genes: int, optional
Exclude GO terms with fewer than this number of genes.
max_genes: int, optional
Exclude GO terms with more than this number of genes.
Returns
-------
GeneSetCollection
A gene set "database" with one gene set for each GO term.
"""
if not self.terms:
raise ValueError('You need to first parse both an OBO file and '
'a gene association file!')
if not self.annotations:
raise ValueError('You need to first parse a gene association '
'file!')
all_term_ids = sorted(self.terms.keys())
# go over all GO terms and get associated genes
logger.info('Obtaining GO term associations...')
# n = len(all_term_ids)
# term_gene_counts = []
# term_ids = []
term_genes = OrderedDict()
geneset_terms = {}
gene_sets = []
for j, id_ in enumerate(all_term_ids):
tg = self.get_goterm_genes(id_)
assert isinstance(tg, frozenset)
c = len(tg)
if c == 0:
continue
if (min_genes is not None and c < min_genes) or \
(max_genes is not None and c > max_genes):
# term doesn't meet min/max number of genes criteria
continue
# for finding redundant terms (use set of genes as key)
try:
geneset_terms[tg].append(id_)
except KeyError:
geneset_terms[tg] = [id_]
term_genes[id_] = tg
selected = len(term_genes)
affected = 0
excl = 0
for id_, tg in term_genes.items():
# check if there are redundant terms
term = self.terms[id_]
if len(geneset_terms[tg]) > 1:
gt = geneset_terms[tg]
affected += 1
# check if this term is an ancestor of any of them
# if so, exclude it
excluded = False
for other_id in gt:
if (other_id != id_) and (other_id in term.descendants):
excluded = True
break
if excluded:
excl += 1
continue
# if the term is not redundant with any other term,
# or if it isn't the ancestor of any redundant term,
# add its gene set to the list
name = term.name
source = 'GO'
coll = term.domain_short
desc = term.definition
gs = GeneSet(id_, name, tg, source=source,
collection=coll, description=desc)
gene_sets.append(gs)
D = GeneSetCollection(gene_sets)
logger.info('# terms selected intially: %d', selected)
logger.info('# terms with redundant gene sets: %d', affected)
logger.info('# terms excluded due to redundancy: %d', excl)
logger.info('# terms retained: %d', D.n)
return D
def test_list(my_gene_set):
l = my_gene_set.to_list()
assert isinstance(l, list)
assert len(l) == 6
other = GeneSet.from_list(l)
assert other == my_gene_set
def test_list(my_gene_set):
l = my_gene_set.to_list()
assert isinstance(l, list)
assert len(l) == 6
other = GeneSet.from_list(l)
assert other == my_gene_set
def my_gene_set2(my_genes):
# a gene set with all optional attributes set to None
gene_set = GeneSet('TestID', 'TestName', my_genes)
return gene_set