def genome_lineages(self):
"""Generate lineage information for genomes.
"""
# identify taxa at standard ranks
ranks = [
'superkingdom', 'kingdom', 'phylum', 'class', 'order', 'family',
'genus', 'species'
]
self.df[ranks[:-1]] = self.df['taxid'].apply(
lambda x: pd.Series(taxids_at_ranks(x, ranks[:-1], self.taxdump)))
# report number of taxa represented at each rank
print('Number of taxonomic groups represented:')
for rank in ranks:
print(' {}: {}.'.format(rank, self.df[rank].nunique()))
# merge superkingdom and kingdom
self.df['kingdom'] = self.df[['superkingdom',
'kingdom']].apply(lambda x: x[1]
if x[1] else x[0],
axis=1)
# generate lineage string
self.df['lineage'] = self.df[ranks[1:]].fillna('').apply(
lambda col: col.apply(lambda val, name: '{}__{}'.format(
name[0], self.taxdump[val]['name'] if val else ''),
args=(col.name, ))).apply('; '.join, axis=1)
# write table
fname = 'lineages.txt'
self.df['lineage'].to_csv(join(self.output, fname),
sep='\t',
header=False)
print('Genome lineages written to {}.'.format(fname))
def test_taxids_at_ranks(self):
taxdump = taxdump_from_text(taxdump_archaea)
ranks = ['phylum', 'class', 'genus', 'species']
obs = taxids_at_ranks('1538547', ranks, taxdump)
exp = {
'phylum': '1655434',
'class': None,
'genus': '1655637',
'species': '1538547'
}
self.assertDictEqual(obs, exp)