def test_mouse_map():
hgnc_id1 = hgnc_client.get_hgnc_from_mouse('109599')
hgnc_id2 = hgnc_client.get_hgnc_from_mouse('MGI:109599')
assert hgnc_id1 == '4820'
assert hgnc_id2 == '4820'
hgnc_id = hgnc_client.get_hgnc_from_mouse('xxx')
assert hgnc_id is None
def mgi_to_hgnc_name(gene_list):
"""Convert given mouse gene symbols to HGNC equivalent symbols"""
filtered_mgi = {
mouse_gene_name_to_mgi[gene]
for gene in gene_list if gene in mouse_gene_name_to_mgi
}
hgnc_gene_set = set()
for mgi_id in filtered_mgi:
hgnc_id = get_hgnc_from_mouse(mgi_id)
hgnc_gene_set.add(get_hgnc_name(hgnc_id))
return hgnc_gene_set
def _refs_from_mgi_id(mgi_id):
ref = {'MGI': mgi_id}
hgnc_id = hgnc_client.get_hgnc_from_mouse(mgi_id)
if hgnc_id is None:
logger.warning('Could not get HGNC ID for MGI ID %s' %
mgi_id)
return None
hgnc_ref = _refs_from_hgnc_id(hgnc_id)
if hgnc_ref is None:
return None
ref.update(hgnc_ref)
return ref
def ligand_mgi_to_hgnc_name(seurat_ligand_genes):
filtered_mgi = defaultdict(set)
for logfc, gene in seurat_ligand_genes.items():
if gene in mouse_gene_name_to_mgi:
filtered_mgi[(gene, logfc)].add(mouse_gene_name_to_mgi[gene])
hgnc_gene_dict = defaultdict(set)
seen_genes = set()
for key, value in filtered_mgi.items():
mgi_id = next(iter(value))
hgnc_id = get_hgnc_from_mouse(mgi_id)
hgnc_symbol = get_hgnc_name(hgnc_id)
if hgnc_symbol not in seen_genes:
hgnc_gene_dict[(key[1])].add(hgnc_symbol)
else:
pass
seen_genes.add(hgnc_symbol)
return hgnc_gene_dict
def mgi_to_hgnc_name(gene_list):
"""Convert given mouse gene symbols to HGNC equivalent symbols"""
mouse_gene_name_to_mgi = {
v: um.uniprot_mgi.get(k)
for k, v in um.uniprot_gene_name.items() if k in um.uniprot_mgi
}
filtered_mgi = {
mouse_gene_name_to_mgi[gene]
for gene in gene_list if gene in mouse_gene_name_to_mgi
}
if len(filtered_mgi) == 0:
return 'None'
#raise Exception('No genes found')
hgnc_gene_set = dict()
for mgi_id in filtered_mgi:
hgnc_id = get_hgnc_from_mouse(mgi_id)
hgnc_name = get_hgnc_name(hgnc_id)
return hgnc_name
def load_mouse_genes(fname):
"""Return a list of human genes based on a table of mouse genes."""
# assumes the csv has headers
df = pandas.read_csv(fname)
for c in df.columns:
# assumes the first column starting with MGI is the relevant one
# with MGI:IDs
if c.startswith('MGI'):
df = df.rename(columns={c: 'MGI'})
break
mgi_ids = df['MGI']
genes = []
for mgi_id in mgi_ids:
if mgi_id.startswith('MGI:'):
mgi_id = mgi_id[4:]
hgnc_id = hgnc_client.get_hgnc_from_mouse(mgi_id)
if not hgnc_id:
print('Could not find human gene corresponding to MGI %s' % mgi_id)
continue
genes.append(hgnc_id)
return genes
def mouse_human_mappings(df):
site_data = df[['MgiId', 'MotifPeptide']].values
human_peptides = []
for mgi_id_str, peptide in site_data:
# Remove --- indicating gaps (start/end of protein)
remove_gap = peptide.replace('-', '')
star_pos = remove_gap.find('*')
# If there's no asterisk (think this happens once in whole dataset)
# skip this peptide
if star_pos == -1:
continue
# Remove the star from the peptide
proc_peptide = remove_gap.replace('*', '')
# Get the position of the target residue (star_pos - 1 + 1)
site_pos = star_pos
# Get Uniprot ID(s) for this gene(s)
human_proteins = set()
# Skip peptides with no MGI ID
if mgi_id_str is np.nan:
continue
for mgi_id in mgi_id_str.split('|'):
mgi_id = mgi_id.split(':')[1]
int(mgi_id)
hgnc_id = hgnc_client.get_hgnc_from_mouse(mgi_id)
if hgnc_id is not None:
up_id_hgnc = hgnc_client.get_uniprot_id(hgnc_id)
#gene_sym = hgnc_client.get_hgnc_name(hgnc_id)
if up_id_hgnc is None:
continue
# If there is more than one hgnc->up_id, try both
up_ids = up_id_hgnc.split(',')
for up_id in up_ids:
human_proteins.add(up_id.strip())
if len(human_proteins) > 1:
print("Warning: >1 protein: %s, %s" %
(mgi_id_str, str(human_proteins)))
for human_prot in human_proteins:
human_peptides.append((human_prot, proc_peptide, site_pos))
return human_peptides
