def compute_distances(self, df=None):
if df is None:
df = self.clone_df
if 'alpha' in self.chains:
pw_alpha = _pws(df=df,
metrics=self.metrics_a,
weights=self.weights_a,
kargs=self.kargs_a,
cpu=self.cpus,
store=self.store_all_cdr)
self._assign_distance_attributes(d=pw_alpha, chain='alpha')
if 'beta' in self.chains:
pw_beta = _pws(df=df,
metrics=self.metrics_b,
weights=self.weights_b,
kargs=self.kargs_b,
cpu=self.cpus,
store=self.store_all_cdr)
self._assign_distance_attributes(d=pw_beta, chain='beta')
if 'gamma' in self.chains:
pw_gamma = _pws(df=df,
metrics=self.metrics_g,
weights=self.weights_g,
kargs=self.kargs_g,
cpu=self.cpus,
store=self.store_all_cdr)
self._assign_distance_attributes(d=pw_gamma, chain='gamma')
if 'delta' in self.chains:
pw_delta = _pws(df=df,
metrics=self.metrics_d,
weights=self.weights_d,
kargs=self.kargs_d,
cpu=self.cpus,
store=self.store_all_cdr)
self._assign_distance_attributes(d=pw_delta, chain='delta')
def test_example_7():
"""
If you don't want to use OOP, but you I still want a multi-CDR
tcrdistances on a single chain, using you own metric
def my_own_metric(s1,s2):
return Levenshtein.distance(s1,s2)
"""
import multiprocessing
import pandas as pd
from tcrdist.rep_funcs import _pws, _pw
df = pd.read_csv("dash2.csv")
metrics_b = {
"cdr3_b_aa": my_own_metric,
"pmhc_b_aa": my_own_metric,
"cdr2_b_aa": my_own_metric,
"cdr1_b_aa": my_own_metric
}
weights_b = {
"cdr3_b_aa": 1,
"pmhc_b_aa": 1,
"cdr2_b_aa": 1,
"cdr1_b_aa": 1
}
kargs_b = {
'cdr3_b_aa': {
'use_numba': False
},
'pmhc_b_aa': {
'use_numba': False
},
'cdr2_b_aa': {
'use_numba': False
},
'cdr1_b_aa': {
'use_numba': False
}
}
dmats = _pws(df=df,
metrics=metrics_b,
weights=weights_b,
kargs=kargs_b,
cpu=1,
uniquify=True,
store=True)
print(dmats.keys())
def test_example_8():
"""
You want a 'tcrdistance' but you don't want to bother with the tcrdist3 framework.
Note that the columns names are completely arbitrary under this
framework, so one can directly compute a tcrdist on a
AIRR, MIXCR, VDJTools, or other formated file without any
reformatting.
"""
import multiprocessing
import pandas as pd
import pwseqdist as pw
from tcrdist.rep_funcs import _pws, _pw
df_airr = pd.read_csv("dash_beta_airr.csv")
# Choose the metrics you want to apply to each CDR
metrics = {
'cdr3_aa': pw.metrics.nb_vector_tcrdist,
'cdr2_aa': pw.metrics.nb_vector_tcrdist,
'cdr1_aa': pw.metrics.nb_vector_tcrdist
}
# Choose the weights that are right for you.
weights = {'cdr3_aa': 3, 'cdr2_aa': 1, 'cdr1_aa': 1}
# Provide arguments for the distance metrics
kargs = {
'cdr3_aa': {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': False
},
'cdr2_aa': {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
'cdr1_aa': {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
}
}
# Here are your distance matrices
from tcrdist.rep_funcs import _pws
dmats = _pws(df=df_airr,
metrics=metrics,
weights=weights,
kargs=kargs,
cpu=1,
store=True)
dmats['tcrdist']
def get_centroid_seq_alpha(df, cdr3_name='cdr3_a_aa'):
"""
Given a list of sequences, returns the sequence with the minimum
sum of distances to all other seqs in the list.
Parameters
----------
seqs : list
list of strings (amino acid rep)
metric : func
defaults to pw.metrics.nw_hamming_metric
Returns
-------
centroid_seq : str
Example
-------
>>> seqs = ['CASSEILAALGTQYF', 'CASSWTSRETQYF', 'CASSLAQETQYF', 'CASSLAPGDVSQYF', 'CASSWDQETQYF', 'CASSLWWDSGANVLTF', 'CASSLARTLSSGANVLTF', 'CASIPGTLFTFSGANVLTF', 'CASSFASSGANVLTF', 'CASSYRLLSGANVLTF']
>>> get_centroid_seq(seqs)
'CASSFASSGANVLTF'
Notes
-----
In case of multiple occurrences of the minimum values, the indices
corresponding to the first occurrence are returned.
"""
#import pwseqdist as pw
#from scipy.spatial.distance import squareform
seqs = df[cdr3_name]
if len(seqs) < 3:
return df.head(1)[cdr3_name], None, None, None
metrics = {
"cdr3_a_aa": pw.metrics.nb_vector_tcrdist,
"pmhc_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr2_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr1_a_aa": pw.metrics.nb_vector_tcrdist
}
# Define weights
weights = {"cdr3_a_aa": 3, "pmhc_a_aa": 1, "cdr2_a_aa": 1, "cdr1_a_aa": 1}
kargs = {
"cdr3_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': False
},
"pmhc_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr2_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr1_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
}
}
dmat = _pws(df=df,
metrics=metrics,
weights=weights,
store=False,
uniquify=False,
kargs=kargs)
dmat = dmat['tcrdist']
dmat = dmat.astype(int)
iloc_idx = dmat.sum(axis=0).argmin()
centroid_seq = df[cdr3_name].to_list()[iloc_idx]
loc_idx = df.index.to_list()[iloc_idx]
return centroid_seq, dmat, iloc_idx, loc_idx
def test_dash_nw_metric_fixed_gappos_False():
import pandas as pd
import pwseqdist as pw
from tcrdist.rep_funcs import _pws
# Define metrics for each region
metrics = {
"cdr3_a_aa": pw.metrics.nw_metric,
"pmhc_a_aa": pw.metrics.nw_metric,
"cdr2_a_aa": pw.metrics.nw_metric,
"cdr1_a_aa": pw.metrics.nw_metric,
"cdr3_b_aa": pw.metrics.nw_metric,
"pmhc_b_aa": pw.metrics.nw_metric,
"cdr2_b_aa": pw.metrics.nw_metric,
"cdr1_b_aa": pw.metrics.nw_metric
}
# Define weights
weights = {
"cdr3_a_aa": 3,
"pmhc_a_aa": 1,
"cdr2_a_aa": 1,
"cdr1_a_aa": 1,
"cdr3_b_aa": 3,
"pmhc_b_aa": 1,
"cdr2_b_aa": 1,
"cdr1_b_aa": 1
}
kargs = {
"cdr3_a_aa": {
'use_numba': False
},
"pmhc_a_aa": {
'use_numba': False
},
"cdr2_a_aa": {
'use_numba': False
},
"cdr1_a_aa": {
'use_numba': False
},
"cdr3_b_aa": {
'use_numba': False
},
"pmhc_b_aa": {
'use_numba': False
},
"cdr2_b_aa": {
'use_numba': False
},
"cdr1_b_aa": {
'use_numba': False
}
}
df = pd.read_csv("dash2.csv")
import multiprocessing
df = df.head(100)
r = _pws(df=df,
metrics=metrics,
weights=weights,
kargs=kargs,
cpu=1,
store=False)
assert r['tcrdist'].shape[0] == 100
assert r['tcrdist'].shape[1] == 100
def test_dash_tcrdist_fixed_gappos_True():
import pandas as pd
import pwseqdist as pw
from tcrdist.rep_funcs import _pws
# Define metrics for each region
metrics = {
"cdr3_a_aa": pw.metrics.nb_vector_tcrdist,
"pmhc_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr2_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr1_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr3_b_aa": pw.metrics.nb_vector_tcrdist,
"pmhc_b_aa": pw.metrics.nb_vector_tcrdist,
"cdr2_b_aa": pw.metrics.nb_vector_tcrdist,
"cdr1_b_aa": pw.metrics.nb_vector_tcrdist
}
# Define weights
weights = {
"cdr3_a_aa": 3,
"pmhc_a_aa": 1,
"cdr2_a_aa": 1,
"cdr1_a_aa": 1,
"cdr3_b_aa": 3,
"pmhc_b_aa": 1,
"cdr2_b_aa": 1,
"cdr1_b_aa": 1
}
kargs = {
"cdr3_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': True
},
"pmhc_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr2_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr1_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr3_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': True
},
"pmhc_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr2_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr1_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
}
}
df = pd.read_csv("dash2.csv")
r = _pws(df=df,
metrics=metrics,
weights=weights,
kargs=kargs,
cpu=1,
store=False)
assert r['tcrdist'].shape[0] == 1924
assert r['tcrdist'].shape[1] == 1924
def test_pws_rectangular_computation():
metrics = {
"cdr3_a_aa": pw.metrics.nb_vector_tcrdist,
"pmhc_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr2_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr1_a_aa": pw.metrics.nb_vector_tcrdist,
"cdr3_b_aa": pw.metrics.nb_vector_tcrdist,
"pmhc_b_aa": pw.metrics.nb_vector_tcrdist,
"cdr2_b_aa": pw.metrics.nb_vector_tcrdist,
"cdr1_b_aa": pw.metrics.nb_vector_tcrdist
}
weights = {
"cdr3_a_aa": 3,
"pmhc_a_aa": 1,
"cdr2_a_aa": 1,
"cdr1_a_aa": 1,
"cdr3_b_aa": 3,
"pmhc_b_aa": 1,
"cdr2_b_aa": 1,
"cdr1_b_aa": 1
}
kargs = {
"cdr3_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': False
},
"pmhc_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr2_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr1_a_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr3_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 3,
'ctrim': 2,
'fixed_gappos': False
},
"pmhc_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr2_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
},
"cdr1_b_aa": {
'use_numba': True,
'distance_matrix': pw.matrices.tcr_nb_distance_matrix,
'dist_weight': 1,
'gap_penalty': 4,
'ntrim': 0,
'ctrim': 0,
'fixed_gappos': True
}
}
df = pd.read_csv("dash2.csv")
df = df.head(10).copy()
df2 = pd.read_csv("dash2.csv")
r = _pws(df=df,
df2=df2,
metrics=metrics,
weights=weights,
kargs=kargs,
cpu=1,
store=False)
assert r['tcrdist'].shape == (10, 1924)