def test_TCRsampler_build():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background()
assert isinstance(t.ref_dict, dict)
assert isinstance(t.ref_dict.popitem()[1], pd.DataFrame)
def test_prob_sampler_sample_key_warn():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background()
with pytest.warns(None):
r = t.sample([['TRBV999*01', 'TRBJ2-7*01', 2]])
assert r == [[None]]
def test_prob_sampler_sample_background():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background()
r = t.sample_background('TRBV9*01', 'TRBJ2-7*01', n=10)
assert r == [
'CASSRTGSLADEQYF', 'CASSATGVVSAQYF', 'CASSAWGQVYEQYF',
'CASSVSGSPYEQYF', 'CASSAWGQVYEQYF', 'CASSAWGQVYEQYF', 'CASRWGEQYF',
'CASSGDDWEQYF', 'CASSATGTSGPYEQYF', 'CASSSRTSGSNSEQYF'
]
def test_TCRsampler_build_vj_components():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background()
assert np.isclose(np.sum([k for _, k in t.vj_freq.items()]), 1.0)
assert np.isclose(np.sum([k for _, k in t.j_freq.items()]), 1.0)
assert np.isclose(np.sum([k for _, k in t.v_freq.items()]), 1.0)
assert np.isclose(np.sum([k for _, k in t.vj_occur_freq.items()]), 1.0)
assert np.isclose(np.sum([k for _, k in t.v_occur_freq.items()]), 1.0)
assert np.isclose(np.sum([k for _, k in t.j_occur_freq.items()]), 1.0)
def test_prob_sampler_sample():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background()
r = t.sample([['TRBV9*01', 'TRBJ2-7*01', 2]])
assert r == [['CASSRTGSLADEQYF', 'CASSATGVVSAQYF']]
r = t.sample([['TRBV9*01', 'TRBJ2-7*01', 2]], flatten=True)
assert r == ['CASSRTGSLADEQYF', 'CASSATGVVSAQYF']
r = t.sample([['TRBV9*01', 'TRBJ2-7*01', 2],
['TRBV7-7*01', 'TRBJ2-4*01', 4]])
assert r == [['CASSRTGSLADEQYF', 'CASSATGVVSAQYF'],
[
'CASSLGQAARGIQYF', 'CASSLGQAARGIQYF', 'CASSLGQAARGIQYF',
'CASSLGQAARGIQYF'
]]
mixcr exportClones -cloneId -count -fraction -vGene -jGene -vHit -jHit -vHits -jHits -aaFeature CDR3 -nFeature CDR3 SRR2079522.1.clns SRR2079522.1.clns.best.txt -f
mixcr exportAlignments SRR2079522.1.vdjca SRR2079522.1.vdjca.txt -f
```
#### Files Available For Download
Beta: [SRR2079522.1.clns.best.txt](https://www.dropbox.com/s/czcewp7x7auwdsu/SRR2079522.1.clns.best.txt?dl=1)
Alpha: [SRR2079521.1.clns.best.txt](https://www.dropbox.com/s/k4i0mt0cwhcn1h7/SRR2079521.1.clns.best.txt?dl=1)
"""
from tcrsampler.sampler import TCRsampler
fn = 'SRR2079522.1.clns.best.subject.txt'
t = TCRsampler()
t.clean_mixcr(fn)
t.build_background()
t.ref_df
t.ref_df.to_csv('ruggiero_mouse_beta_t.tsv.sampler.tsv', sep="\t", index=False)
fn = 'SRR2079521.1.clns.best.subject.txt'
t = TCRsampler()
t.clean_mixcr(fn)
t.build_background()
t.ref_df
t.ref_df.to_csv('ruggiero_mouse_alpha_t.tsv.sampler.tsv',
sep="\t",
index=False)
df['strain'] = 'C57BL6 inbred mouse strain'
print(df)
wirasinha = pd.read_csv(
'/Volumes/Samsung_T5/kmayerbl/tcr_data/wirasinha/Wirasinha.migec.txt',
sep='\t')
for i, row in df.iterrows():
sdf = subset_wirasinha(df=wirasinha,
subset=row['subset'],
tcr_b=row['tcr_b'],
chain=row['chain'])
sdf[['bestv',
'bestj']] = sdf[['v', 'j']].apply(lambda x: x.apply(_pick_best))
sdf[['bestv',
'bestj']] = sdf[['bestv',
'bestj']].apply(lambda x: x.apply(_strip_allele))
sdf = sdf.rename(columns=wirasinha_to_mixcr_headers)
sys.stdout.write(f"Writing {row['filename']}\n")
sdf.to_csv(row['filename'], sep="\t")
sys.stdout.write(
f"Testing {row['filename']} for import into TCRsampler\t")
t = TCRsampler()
t.clean_mixcr(filename=row['filename'])
t.build_background()
print("\n")
print(t.ref_df.head(3))
name = f"{row['filename']}.sampler.tsv"
sys.stdout.write(f"Writing {name} \t")
t.ref_df.to_csv(name, sep="\t", index=False)
def test_TCRsampler_build_stratified():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
t.build_background(stratify_by_subject=True)
r = t.sample_background('TRBV9*01', 'TRBJ2-7*01', n=10)
def test_TCRsampler_clean_mixcr():
t = TCRsampler()
fn = os.path.join('tcrsampler', 'tests', 'pmbc_mixcr_example_data.txt')
t.clean_mixcr(filename=fn)
assert isinstance(t.ref_df, pd.DataFrame)