def getComponents(matches, max_distance=0, min_overlap=0, by_query=False):
"""return overlapping matches.
max_distance
allow reads to be joined if they are # residues apart.
Adjacent reads are 1 residue apart, overlapping reads are 0 residues
apart
min_overlap
require at least # residues to be overlapping
"""
addAlignments(matches, by_query=by_query)
components = Components.IComponents()
for x in range(0, len(matches)):
components.add(x, x)
if min_overlap > 0 and max_distance > 0:
raise ValueError(
"both min_overlap (%i) and max_distance (%i) > 0" % (min_overlap, max_distance))
if by_query:
if min_overlap > 0:
f = lambda x, y: alignlib_lite.py_getAlignmentOverlap(
matches[x].mMapQuery2Target,
matches[y].mMapQuery2Target,
alignlib_lite.py_RR) >= min_overlap
else:
f = lambda x, y: alignlib_lite.py_getAlignmentShortestDistance(
matches[x].mMapQuery2Target,
matches[y].mMapQuery2Target,
alignlib_lite.py_RR) <= max_distance
else:
if min_overlap > 0:
f = lambda x, y: alignlib_lite.py_getAlignmentOverlap(
matches[x].mMapTarget2Query,
matches[y].mMapTarget2Query,
alignlib_lite.py_RR) >= min_overlap
else:
f = lambda x, y: alignlib_lite.py_getAlignmentShortestDistance(
matches[x].mMapTarget2Query,
matches[y].mMapTarget2Query,
alignlib_lite.py_RR) <= max_distance
for x in range(len(matches)):
for y in range(0, x):
if f(x, y):
components.add(x, y)
return components.getComponents()
def MapTranscripts2Genes(transcripts, map_transcript2location):
"""map all orthologous and overlapping transcripts into genes.
The new gene is chosen at random.
"""
graph = Components.SComponents()
map_id2info = {}
added = set()
for transcript in transcripts:
map_id2info[transcript.mTranscript] = (
transcript.mSchema, transcript.mQuality)
token1, strand1, from1, to1 = map_transcript2location[
transcript.mTranscript]
# add link to self as otherwise the component is empty
graph.add(transcript.mTranscript, transcript.mTranscript)
# add link to overlapping transcripts
for x in added:
token2, strand2, from2, to2 = map_transcript2location[x]
if token1 == token2 and strand1 == strand2 and \
min(to1, to2) - max(from1, from2) > 0:
graph.add(transcript.mTranscript, x)
added.add(transcript.mTranscript)
components = graph.getComponents()
new_genes = {}
new_transcripts = []
for component in components:
g = component[0]
new_genes[g] = []
for id in component:
s, q = map_id2info[id]
t = Orthologs.Transcript()
t.mSchema = s
t.mTranscript = id
t.mGene = g
t.mQuality = q
new_genes[g].append(t)
new_transcripts.append(t)
return new_transcripts, new_genes
import CGAT.Components as Components
c = Components.SComponents()
links = (("1", "2"), ("1", "3"), ("2", "3"), ("3", "3"), ("4", "5"), ("6",
"6"))
for a, b in links:
print a, b, c.add(a, b)
for x in "01234567":
print x, c.get(x)
print c.getNumNodes()
print c.getComponents()
c = Components.IComponents()
for a, b in links:
print a, b, c.add(int(a), int(b))
for x in range(0, 8):
print x, c.get(x)
print c.getNumNodes()
print c.getComponents()
c = Components.IComponents()
print c.getComponents()