| 1 |
""" |
| 2 |
DSSP Parser |
| 3 |
""" |
| 4 |
|
| 5 |
import csb.core |
| 6 |
import csb.io |
| 7 |
|
| 8 |
from csb.bio.structure import SecStructures, UnknownSecStructureError |
| 9 |
|
| 10 |
|
| 11 |
class DSSPParseError(ValueError): |
| 12 |
pass |
| 13 |
|
| 14 |
|
| 15 |
class ResidueAssignmentInfo(object): |
| 16 |
|
| 17 |
def __init__(self, residue_id, accession, chain, secondary_structure, phi, psi): |
| 18 |
|
| 19 |
self.residue_id = residue_id |
| 20 |
self.accession = accession |
| 21 |
self.chain = chain |
| 22 |
self.secondary_structure = secondary_structure |
| 23 |
self.phi = phi |
| 24 |
self.psi = psi |
| 25 |
|
| 26 |
|
| 27 |
class DSSPParser(object): |
| 28 |
""" |
| 29 |
Simple DSSP Secondary Structure Parser. |
| 30 |
""" |
| 31 |
|
| 32 |
def parse(self, dssp_file): |
| 33 |
""" |
| 34 |
@param dssp_file: source DSSP file to parse |
| 35 |
@type dssp_file: str |
| 36 |
@return: a dictionary of L{ResidueAssignmentInfo} objects |
| 37 |
@rtype: dict |
| 38 |
""" |
| 39 |
|
| 40 |
data = {} |
| 41 |
start = False |
| 42 |
offset = 0 # assume old DSSP format |
| 43 |
accession = None |
| 44 |
|
| 45 |
for line in open(dssp_file): |
| 46 |
|
| 47 |
if not start: |
| 48 |
|
| 49 |
if line.startswith('HEADER'): |
| 50 |
accession = line[62:66].strip().lower() |
| 51 |
|
| 52 |
elif line.startswith(' # RESIDUE'): |
| 53 |
if len(line) >= 140: |
| 54 |
offset = 4 # the new DSSP format |
| 55 |
start = True |
| 56 |
else: |
| 57 |
if line[13] == '!': |
| 58 |
continue |
| 59 |
|
| 60 |
residue_id = line[6:11].strip() |
| 61 |
chain = line[11] |
| 62 |
try: |
| 63 |
ss = line[16].strip() |
| 64 |
if ss == '': |
| 65 |
ss = SecStructures.Gap |
| 66 |
else: |
| 67 |
ss = csb.core.Enum.parse(SecStructures, ss) |
| 68 |
except csb.core.EnumValueError as e: |
| 69 |
raise UnknownSecStructureError(str(e)) |
| 70 |
phi = float(line[104 + offset : 109 + offset]) |
| 71 |
psi = float(line[110 + offset : 115 + offset]) |
| 72 |
|
| 73 |
|
| 74 |
if chain not in data: |
| 75 |
data[chain] = {} |
| 76 |
|
| 77 |
data[chain][residue_id] = ResidueAssignmentInfo(residue_id, accession, chain, ss, phi, psi) |
| 78 |
|
| 79 |
return data |
| 80 |
|
| 81 |
class StrideParser(object): |
| 82 |
""" |
| 83 |
Simple STRIDE Secondary Structure Parser. |
| 84 |
""" |
| 85 |
|
| 86 |
def parse(self, stride_file): |
| 87 |
""" |
| 88 |
@param stride_file: source STRIDE file to parse |
| 89 |
@type stride_file: str |
| 90 |
@return: a dictionary of L{ResidueAssignmentInfo} objects |
| 91 |
@rtype: dict |
| 92 |
""" |
| 93 |
|
| 94 |
data = {} |
| 95 |
|
| 96 |
for line in open(stride_file): |
| 97 |
if line.startswith('ASG '): |
| 98 |
|
| 99 |
fields = line.split() |
| 100 |
|
| 101 |
residue_id = fields[3] |
| 102 |
chain = fields[2] |
| 103 |
accession = fields[-1].lower() |
| 104 |
try: |
| 105 |
ss = csb.core.Enum.parse(SecStructures, fields[5]) |
| 106 |
except csb.core.EnumValueError as e: |
| 107 |
raise UnknownSecStructureError(str(e)) |
| 108 |
phi = float(fields[7]) |
| 109 |
psi = float(fields[8]) |
| 110 |
|
| 111 |
if chain not in data: |
| 112 |
data[chain] = {} |
| 113 |
|
| 114 |
data[chain][residue_id] = ResidueAssignmentInfo(residue_id, accession, chain, ss, phi, psi) |
| 115 |
|
| 116 |
return data |
| 117 |
|
| 118 |
|
| 119 |
def get(accession, prefix='http://www.pdb.org/pdb/files/'): |
| 120 |
""" |
| 121 |
Download and parse a DSSP entry. |
| 122 |
|
| 123 |
@param accession: accession number of the entry |
| 124 |
@type accession: str |
| 125 |
@param prefix: download URL prefix |
| 126 |
@type prefix: str |
| 127 |
|
| 128 |
@return: see L{DSSPParser.parse} |
| 129 |
@rtype: dict |
| 130 |
""" |
| 131 |
dssp = csb.io.TempFile() |
| 132 |
|
| 133 |
browser = csb.io.urllib.urlopen(prefix + accession.lower() + '.dssp') |
| 134 |
dssp.write(browser.read().decode('utf-8')) |
| 135 |
dssp.flush() |
| 136 |
|
| 137 |
return DSSPParser().parse(dssp.name) |
| 138 |
|