def test_parse2(self):
os.chdir(
self.thisd
) # Need as otherwise tests that happen in other directories change os.cwd()
dsspLog = os.path.join(self.testfiles_dir, "3ouf.dssp")
dsspP = dssp_parser.DsspParser(dsspLog)
def test_parse1(self):
os.chdir(
self.thisd
) # Need as otherwise tests that happen in other directories change os.cwd()
dsspLog = os.path.join(self.testfiles_dir, "2bhw.dssp")
dsspP = dssp_parser.DsspParser(dsspLog)
a = [
' ', ' ', 'T', 'T', ' ', 'T', 'T', 'S', 'S', 'T', 'T', ' ', ' ',
' ', 'T', 'T', 'G', 'G', 'G', ' ', ' ', 'S', ' ', ' ', 'T', 'T',
' ', ' ', 'S', ' ', 'S', 'T', 'T', ' ', ' ', 'S', ' ', ' ', 'T',
'T', ' ', 'T', 'T', ' ', 'S', 'H', 'H', 'H', 'H', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'T',
'T', 'T', ' ', ' ', ' ', 'S', ' ', ' ', 'S', 'G', 'G', 'G', 'S',
'G', 'G', 'G', 'G', 'G', 'S', 'T', 'T', ' ', 'E', 'E', 'G', 'G',
'G', ' ', 'T', 'T', 'S', 'E', 'E', 'E', ' ', ' ', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'E', 'E', 'T', 'T', 'E', 'E', ' ', 'S',
'S', 'S', 'S', ' ', ' ', ' ', 'T', 'T', 'S', ' ', 'T', 'T', ' ',
'T', 'T', ' ', 'S', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H',
'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', ' ', 'S', ' ',
'S', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', 'H', ' ', 'T', 'T',
'T', 'S', 'S', 'G', 'G', 'G', 'G', 'T', 'T', 'T', 'T', ' ', 'S',
' ', ' '
]
self.assertEqual(dsspP.assignment[0], a)
def helixFromContacts(self, contacts, dsspLog, minContig=2, maxGap=3):
"""Return the sequence of the longest contiguous helix from the given contact data
source is the native, target the model
Get start and stop indices of all contiguous chunks
- we can match multiple chains in the model, but any match must be within a single native chain
- the source can increment or decrement - the model only ever increments (ncont "sort target inc" argument)
- matches can be in-register or out-of-register, but for finding the longest chunk we don't care
startstop = [ (10,15), (17, 34), (38, 50) ]
# Loop through all chunks, and if any two have a gap of < mingap, join the indices together.
# Get the indices of the largest chunk
# Get the corresponding AA sequence
"""
if contacts is None or not len(contacts):
return None
dsspP = dssp_parser.DsspParser(dsspLog)
#
# Loop through the contacts finding the start, stop indices in the list of contacts of contiguous chunks
#
chunks = self.findChunks(contacts=contacts,
dsspP=dsspP,
ssTest=True,
minContig=2)
if not chunks:
return None
#
# Go through the start-stop chunks in pairs and see if they can be joined, creating
# extended, which is the list of chunks with gaps filled-in
#
if len(chunks) > 1:
# Need to sort the chunks by chain and then startResSeq so that we can join anything on the same chain
chunks.sort(key=itemgetter('chainId1', 'startResSeq')) # By chain
extended = []
for i, newChunk in enumerate(chunks):
# initialise
if i == 0:
toJoin = newChunk
continue
chunk, toJoin = self._join_chunks(toJoin,
newChunk,
dsspP=dsspP,
maxGap=maxGap)
if chunk is not None:
extended.append(chunk)
# Last one needs to be handled specially
if i == len(chunks) - 1 and toJoin:
extended.append(toJoin)
#
# Find the biggest
#
biggest = sorted(
extended,
lambda x, y: abs(x['stopResSeq'] - x['startResSeq']) - abs(y[
'stopResSeq'] - y['startResSeq']),
reverse=True)[0]
else:
biggest = chunks[0]
#
# Get the sequence that the start, stop indices define
#
chainId = biggest['chainId1']
startResSeq = min(
biggest['startResSeq'],
biggest['stopResSeq']) # use min/max as could be running backwards
stopResSeq = max(biggest['startResSeq'], biggest['stopResSeq'])
sequence = ""
for resSeq in range(startResSeq, stopResSeq + 1):
resName = dsspP.getResName(resSeq, chainId)
sequence += resName
return sequence