def countUMINaive(molecular_barcodes, allowed_mismatches):
"""
Tries to finds clusters of similar UMIs using a naive proximity
approach where UMIs are sorted and the ones that are consecutive
and has hamming distance below the given number of miss-matches will
be clustered together.
It returns a list with all the non clustered UMIs, for clusters of
multiple UMIs a random one will be selected.
:param molecular_barcodes: a list of UMIs
:param allowed_mismatches: how much distance we allow between clusters
:param method: the type of distance algorithm when clustering
(single more restrictive or complete less restrictive)
:type allowed_mismatches: integer
:type method: str
:return: a list of unique UMIs
:rtype: list
"""
clusters_dict = {}
nclusters = 0
for i, molecular_barcode in enumerate(sorted(molecular_barcodes)):
if i == 0:
clusters_dict[nclusters] = [molecular_barcode]
else:
# compare distant of previous molecular barcodes and new one
# if distance is between threshold we add it to the cluster
# otherwise we create a new cluster
if hamming_distance(clusters_dict[nclusters][-1], molecular_barcode) <= allowed_mismatches:
clusters_dict[nclusters].append(molecular_barcode)
else:
nclusters += 1
clusters_dict[nclusters] = [molecular_barcode]
# Return the non clustered UMIs
return [random.choice(members) for members in clusters_dict.itervalues()]
def countUMINaive(molecular_barcodes, allowed_mismatches):
"""
Tries to finds clusters of similar UMIs using a naive proximity
approach where UMIs are sorted and the ones that are consecutive
and has hamming distance below the given number of miss-matches will
be clustered together.
It returns a list with all the non clustered UMIs, for clusters of
multiple UMIs a random one will be selected.
:param molecular_barcodes: a list of UMIs
:param allowed_mismatches: how much distance we allow between clusters
:param method: the type of distance algorithm when clustering
(single more restrictive or complete less restrictive)
:type allowed_mismatches: integer
:type method: str
:return: a list of unique UMIs
:rtype: list
"""
clusters_dict = {}
nclusters = 0
for i, molecular_barcode in enumerate(sorted(molecular_barcodes)):
if i == 0:
clusters_dict[nclusters] = [molecular_barcode]
else:
# compare distant of previous molecular barcodes and new one
# if distance is between threshold we add it to the cluster
# otherwise we create a new cluster
if hamming_distance(clusters_dict[nclusters][-1],
molecular_barcode) <= allowed_mismatches:
clusters_dict[nclusters].append(molecular_barcode)
else:
nclusters += 1
clusters_dict[nclusters] = [molecular_barcode]
# Return the non clustered UMIs
return [random.choice(members) for members in clusters_dict.itervalues()]
def get_adj_list_directional_adjacency(umis, counts):
return {
umi: [
umi2 for umi2 in umis
if hamming_distance(umi, umi2) <= allowed_mismatches
and counts[umi] >= (counts[umi2] * 2) - 1
]
for umi in umis
}
def affinity_umi_removal(molecular_barcodes, _):
"""
Tries to finds clusters of similar UMIs using an affinity based approach.
It returns a list with all the non clustered UMIs, for clusters of
multiple UMIs a random one will be selected.
:param molecular_barcodes: a list of UMIs
:return: a list of unique UMIs
:rtype: list
"""
if len(molecular_barcodes) <= 2:
return countUMINaive(molecular_barcodes, allowed_mismatches)
words = np.asarray(molecular_barcodes)
lev_similarity = -1 * np.array([[hamming_distance(w1,w2) for w1 in words] for w2 in words])
affprop = AffinityPropagation(affinity="precomputed", damping=0.5)
affprop.fit(lev_similarity)
unique_clusters = list()
for cluster_id in np.unique(affprop.labels_):
exemplar = words[affprop.cluster_centers_indices_[cluster_id]]
cluster = np.unique(words[np.nonzero(affprop.labels_==cluster_id)])
unique_clusters.append(random.choice(cluster))
return unique_clusters
def d(coord):
i, j = coord
return hamming_distance(molecular_barcodes[i], molecular_barcodes[j])
def get_adj_list_adjacency(umis):
return {umi: [umi2 for umi2 in umis if hamming_distance(umi, umi2) \
<= allowed_mismatches] for umi in umis}
def d(coord):
i,j = coord
return hamming_distance(molecular_barcodes[i], molecular_barcodes[j])
def get_adj_list_directional_adjacency(umis, counts):
return {umi: [umi2 for umi2 in umis if hamming_distance(umi, umi2) <= allowed_mismatches and
counts[umi] >= (counts[umi2]*2)-1] for umi in umis}
def get_adj_list_adjacency(umis):
return {umi: [umi2 for umi2 in umis if hamming_distance(umi, umi2) \
<= allowed_mismatches] for umi in umis}
def d(coord):
i, j = coord
return hamming_distance(molecular_barcodes[i].encode("UTF-8"),
molecular_barcodes[j].encode("UTF-8"))
def get_adj_list_adjacency(umis):
return {umi: [umi2 for umi2 in umis if hamming_distance(umi.encode("UTF-8"),
umi2.encode("UTF-8")) \
<= allowed_mismatches] for umi in umis}