Upload utils.py
Browse files
utils.py
ADDED
|
@@ -0,0 +1,403 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Copyright 2021 Gabriele Orlando
|
| 2 |
+
#
|
| 3 |
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
| 4 |
+
# you may not use this file except in compliance with the License.
|
| 5 |
+
# You may obtain a copy of the License at
|
| 6 |
+
#
|
| 7 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
| 8 |
+
#
|
| 9 |
+
# Unless required by applicable law or agreed to in writing, software
|
| 10 |
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
| 11 |
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 12 |
+
# See the License for the specific language governing permissions and
|
| 13 |
+
# limitations under the License.
|
| 14 |
+
|
| 15 |
+
import os,torch
|
| 16 |
+
from pyuul.sources.globalVariables import *
|
| 17 |
+
from pyuul.sources import hashings
|
| 18 |
+
|
| 19 |
+
import numpy as np
|
| 20 |
+
import random
|
| 21 |
+
|
| 22 |
+
def setup_seed(seed):
|
| 23 |
+
torch.manual_seed(seed)
|
| 24 |
+
torch.cuda.manual_seed_all(seed)
|
| 25 |
+
np.random.seed(seed)
|
| 26 |
+
random.seed(seed)
|
| 27 |
+
torch.backends.cudnn.deterministic = True
|
| 28 |
+
setup_seed(100)
|
| 29 |
+
|
| 30 |
+
def parseSDF(SDFFile):
|
| 31 |
+
"""
|
| 32 |
+
function to parse pdb files. It can be used to parse a single file or all the pdb files in a folder. In case a folder is given, the coordinates are gonna be padded
|
| 33 |
+
|
| 34 |
+
Parameters
|
| 35 |
+
----------
|
| 36 |
+
SDFFile : str
|
| 37 |
+
path of the PDB file or of the folder containing multiple PDB files
|
| 38 |
+
|
| 39 |
+
Returns
|
| 40 |
+
-------
|
| 41 |
+
coords : torch.Tensor
|
| 42 |
+
coordinates of the atoms in the pdb file(s). Shape ( batch, numberOfAtoms, 3)
|
| 43 |
+
|
| 44 |
+
atomNames : list
|
| 45 |
+
a list of the atom identifier. It encodes atom type, residue type, residue position and chain
|
| 46 |
+
|
| 47 |
+
"""
|
| 48 |
+
if not os.path.isdir(SDFFile):
|
| 49 |
+
fil = SDFFile
|
| 50 |
+
totcoords=[]
|
| 51 |
+
totaname=[]
|
| 52 |
+
coords = []
|
| 53 |
+
atomNames = []
|
| 54 |
+
for line in open(fil).readlines():
|
| 55 |
+
a=line.strip().split()
|
| 56 |
+
if len(a)==16: ## atom
|
| 57 |
+
element = a[3]
|
| 58 |
+
x = float(a[0])
|
| 59 |
+
y = float(a[1])
|
| 60 |
+
z = float(a[2])
|
| 61 |
+
coords += [[x,y,z]]
|
| 62 |
+
#aname = line[17:20].strip()+"_"+str(resnum)+"_"+line[12:16].strip()+"_"+line[21]
|
| 63 |
+
aname = "MOL"+"_"+"0"+"_"+element+"_"+"A"
|
| 64 |
+
|
| 65 |
+
atomNames += [aname]
|
| 66 |
+
elif "$$$$" in line:
|
| 67 |
+
totcoords+=[torch.tensor(coords)]
|
| 68 |
+
totaname += [atomNames]
|
| 69 |
+
coords=[]
|
| 70 |
+
atomNames=[]
|
| 71 |
+
return torch.torch.nn.utils.rnn.pad_sequence(totcoords, batch_first=True, padding_value=PADDING_INDEX),totaname
|
| 72 |
+
else:
|
| 73 |
+
totcoords = []
|
| 74 |
+
totaname = []
|
| 75 |
+
for fil in sorted(os.listdir(SDFFile)):
|
| 76 |
+
coords = []
|
| 77 |
+
atomNames = []
|
| 78 |
+
for line in open(SDFFile+fil).readlines():
|
| 79 |
+
a = line.strip().split()
|
| 80 |
+
if len(a) == 16: ## atom
|
| 81 |
+
element = a[3]
|
| 82 |
+
x = float(a[0])
|
| 83 |
+
y = float(a[1])
|
| 84 |
+
z = float(a[2])
|
| 85 |
+
coords += [[x, y, z]]
|
| 86 |
+
aname = "MOL"+"_"+"0"+"_"+element+"_"+"A"
|
| 87 |
+
|
| 88 |
+
atomNames += [aname]
|
| 89 |
+
elif "$$$$" in line:
|
| 90 |
+
totcoords += [torch.tensor(coords)]
|
| 91 |
+
totaname += [atomNames]
|
| 92 |
+
coords = []
|
| 93 |
+
atomNames = []
|
| 94 |
+
return torch.torch.nn.utils.rnn.pad_sequence(totcoords, batch_first=True, padding_value=PADDING_INDEX),totaname
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
def parsePDB(PDBFile,keep_only_chains=None,keep_hetatm=True,bb_only=False):
|
| 98 |
+
|
| 99 |
+
"""
|
| 100 |
+
function to parse pdb files. It can be used to parse a single file or all the pdb files in a folder. In case a folder is given, the coordinates are gonna be padded
|
| 101 |
+
|
| 102 |
+
Parameters
|
| 103 |
+
----------
|
| 104 |
+
PDBFile : str
|
| 105 |
+
path of the PDB file or of the folder containing multiple PDB files
|
| 106 |
+
bb_only : bool
|
| 107 |
+
if True ignores all the atoms but backbone N, C and CA
|
| 108 |
+
keep_only_chains : str or None
|
| 109 |
+
ignores all the chain but the one given. If None it keeps all chains
|
| 110 |
+
keep_hetatm : bool
|
| 111 |
+
if False it ignores heteroatoms
|
| 112 |
+
Returns
|
| 113 |
+
-------
|
| 114 |
+
coords : torch.Tensor
|
| 115 |
+
coordinates of the atoms in the pdb file(s). Shape ( batch, numberOfAtoms, 3)
|
| 116 |
+
|
| 117 |
+
atomNames : list
|
| 118 |
+
a list of the atom identifier. It encodes atom type, residue type, residue position and chain
|
| 119 |
+
|
| 120 |
+
"""
|
| 121 |
+
|
| 122 |
+
bbatoms = ["N", "CA", "C"]
|
| 123 |
+
if not os.path.isdir(PDBFile):
|
| 124 |
+
fil = PDBFile
|
| 125 |
+
coords = []
|
| 126 |
+
atomNames = []
|
| 127 |
+
cont = -1
|
| 128 |
+
oldres=-999
|
| 129 |
+
for line in open(fil).readlines():
|
| 130 |
+
|
| 131 |
+
|
| 132 |
+
if line[:4] == "ATOM":
|
| 133 |
+
if keep_only_chains is not None and (not line[21] in keep_only_chains):
|
| 134 |
+
continue
|
| 135 |
+
if bb_only and not line[12:16].strip() in bbatoms:
|
| 136 |
+
continue
|
| 137 |
+
if oldres != int(line[22:26]):
|
| 138 |
+
cont+=1
|
| 139 |
+
oldres=int(line[22:26])
|
| 140 |
+
resnum = int(line[22:26])
|
| 141 |
+
atomNames += [line[17:20].strip()+"_"+str(resnum)+"_"+line[12:16].strip()+"_"+line[21]]
|
| 142 |
+
|
| 143 |
+
x = float(line[30:38])
|
| 144 |
+
y = float(line[38:46])
|
| 145 |
+
z = float(line[47:54])
|
| 146 |
+
coords+=[[x,y,z]]
|
| 147 |
+
|
| 148 |
+
elif line[:6] == "HETATM" and keep_hetatm:
|
| 149 |
+
|
| 150 |
+
resname_het = line[17:20].strip()
|
| 151 |
+
resnum = int(line[22:26])
|
| 152 |
+
x = float(line[30:38])
|
| 153 |
+
y = float(line[38:46])
|
| 154 |
+
z = float(line[47:54])
|
| 155 |
+
coords += [[x, y, z]]
|
| 156 |
+
atnameHet = line[12:16].strip()
|
| 157 |
+
atomNames += [resname_het+"_"+str(resnum)+"_"+atnameHet+"_"+line[21]]
|
| 158 |
+
return torch.tensor(coords).unsqueeze(0), [atomNames]
|
| 159 |
+
else:
|
| 160 |
+
coords = []
|
| 161 |
+
atomNames = []
|
| 162 |
+
pdbname = []
|
| 163 |
+
pdb_num = 0
|
| 164 |
+
for fil in sorted(os.listdir(PDBFile)):
|
| 165 |
+
# print(pdb_num)
|
| 166 |
+
pdb_num +=1
|
| 167 |
+
pdbname.append(fil)
|
| 168 |
+
atomNamesTMP = []
|
| 169 |
+
coordsTMP = []
|
| 170 |
+
cont = -1
|
| 171 |
+
oldres=-999
|
| 172 |
+
for line in open(PDBFile+"/"+fil).readlines():
|
| 173 |
+
|
| 174 |
+
if line[:4] == "ATOM":
|
| 175 |
+
if keep_only_chains is not None and (not line[21] in keep_only_chains):
|
| 176 |
+
continue
|
| 177 |
+
if bb_only and not line[12:16].strip() in bbatoms:
|
| 178 |
+
continue
|
| 179 |
+
if oldres != int(line[22:26]):
|
| 180 |
+
cont += 1
|
| 181 |
+
oldres = int(line[22:26])
|
| 182 |
+
|
| 183 |
+
resnum = int(line[22:26])
|
| 184 |
+
atomNamesTMP += [line[17:20].strip()+"_"+str(resnum)+"_"+line[12:16].strip()+"_"+line[21]]
|
| 185 |
+
|
| 186 |
+
x = float(line[30:38])
|
| 187 |
+
y = float(line[38:46])
|
| 188 |
+
z = float(line[47:54])
|
| 189 |
+
coordsTMP+=[[x,y,z]]
|
| 190 |
+
|
| 191 |
+
elif line[:6] == "HETATM" and keep_hetatm:
|
| 192 |
+
if line[17:20].strip()!="GTP":
|
| 193 |
+
continue
|
| 194 |
+
x = float(line[30:38])
|
| 195 |
+
y = float(line[38:46])
|
| 196 |
+
z = float(line[47:54])
|
| 197 |
+
resnum = int(line[22:26])
|
| 198 |
+
coordsTMP += [[x, y, z]]
|
| 199 |
+
atnameHet = line[12:16].strip()
|
| 200 |
+
atomNamesTMP += ["HET_"+str(resnum)+"_"+atnameHet+"_"+line[21]]
|
| 201 |
+
coords+=[torch.tensor(coordsTMP)]
|
| 202 |
+
atomNames += [atomNamesTMP]
|
| 203 |
+
|
| 204 |
+
return torch.torch.nn.utils.rnn.pad_sequence(coords, batch_first=True, padding_value=PADDING_INDEX), atomNames, pdbname, pdb_num
|
| 205 |
+
|
| 206 |
+
|
| 207 |
+
def atomlistToChannels(atomNames, hashing="Element_Hashing", device="cpu"):
|
| 208 |
+
"""
|
| 209 |
+
function to get channels from atom names (obtained parsing the pdb files with the parsePDB function)
|
| 210 |
+
|
| 211 |
+
Parameters
|
| 212 |
+
----------
|
| 213 |
+
atomNames : list
|
| 214 |
+
atom names obtained parsing the pdb files with the parsePDB function
|
| 215 |
+
|
| 216 |
+
hashing : "TPL_Hashing" or "Element_Hashing" or dict
|
| 217 |
+
define which atoms are grouped together. You can use two default hashings or build your own hashing:
|
| 218 |
+
|
| 219 |
+
TPL_Hashing: uses the hashing of torch protein library (https://github.com/lupoglaz/TorchProteinLibrary)
|
| 220 |
+
Element_Hashing: groups atoms in accordnce with the element only: C -> 0, N -> 1, O ->2, P ->3, S- >4, H ->5, everything else ->6
|
| 221 |
+
|
| 222 |
+
Alternatively, if you are not happy with the default hashings, you can build a dictionary of dictionaries that defines the channel of every atom type in the pdb.
|
| 223 |
+
the first dictionary has the residue tag (three letters amino acid code) as key (3 letters compound name for hetero atoms, as written in the PDB file)
|
| 224 |
+
every residue key is associated to a dictionary, which the atom tags (as written in the PDB files) as keys and the channel (int) as value
|
| 225 |
+
|
| 226 |
+
for example, you can define the channels just based on the atom element as following:
|
| 227 |
+
{
|
| 228 |
+
'CYS': {'N': 1, 'O': 2, 'C': 0, 'SG': 3, 'CB': 0, 'CA': 0}, # channels for cysteine atoms
|
| 229 |
+
'GLY': {'N': 1, 'O': 2, 'C': 0, 'CA': 0}, # channels for glycine atom
|
| 230 |
+
...
|
| 231 |
+
'GOL': {'O1':2,'O2':2,'O3':2,'C1':0,'C2':0,'C3':0}, # channels for glycerol atom
|
| 232 |
+
...
|
| 233 |
+
}
|
| 234 |
+
|
| 235 |
+
The default encoding is the one that assigns a different channel to each element
|
| 236 |
+
|
| 237 |
+
other encodings can be found in sources/hashings.py
|
| 238 |
+
|
| 239 |
+
device : torch.device
|
| 240 |
+
The device on which the model should run. E.g. torch.device("cuda") or torch.device("cpu:0")
|
| 241 |
+
Returns
|
| 242 |
+
-------
|
| 243 |
+
coords : torch.Tensor
|
| 244 |
+
coordinates of the atoms in the pdb file(s). Shape ( batch, numberOfAtoms, 3)
|
| 245 |
+
|
| 246 |
+
channels : torch.tensor
|
| 247 |
+
the channel of every atom. Shape (batch,numberOfAtoms)
|
| 248 |
+
|
| 249 |
+
"""
|
| 250 |
+
if hashing == "TPL_Hashing":
|
| 251 |
+
hashing = hashings.TPLatom_hash
|
| 252 |
+
|
| 253 |
+
elif hashing == "Element_Hashing":
|
| 254 |
+
hashing = hashings.elements_hash
|
| 255 |
+
else:
|
| 256 |
+
assert type(hashing) is dict
|
| 257 |
+
|
| 258 |
+
if type(hashing[list(hashing.keys())[0]]) == dict:
|
| 259 |
+
useResName = True
|
| 260 |
+
else:
|
| 261 |
+
useResName = False
|
| 262 |
+
assert type(hashing[list(hashing.keys())[0]]) == int
|
| 263 |
+
channels = []
|
| 264 |
+
for singleAtomList in atomNames:
|
| 265 |
+
haTMP = []
|
| 266 |
+
for i in singleAtomList:
|
| 267 |
+
resname = i.split("_")[0]
|
| 268 |
+
atName = i.split("_")[2]
|
| 269 |
+
# if resname=="HET":
|
| 270 |
+
# atName="HET"
|
| 271 |
+
if useResName:
|
| 272 |
+
if resname in hashing and atName in hashing[resname]:
|
| 273 |
+
haTMP += [hashing[resname][atName]]
|
| 274 |
+
else:
|
| 275 |
+
haTMP += [PADDING_INDEX]
|
| 276 |
+
print("missing ", resname, atName)
|
| 277 |
+
else:
|
| 278 |
+
if atName in hashing:
|
| 279 |
+
haTMP += [hashing[atName]]
|
| 280 |
+
elif atName[0] in hashing:
|
| 281 |
+
haTMP += [hashing[atName[0]]]
|
| 282 |
+
elif hashing == "Element_Hashing":
|
| 283 |
+
haTMP += [6]
|
| 284 |
+
else:
|
| 285 |
+
haTMP += [PADDING_INDEX]
|
| 286 |
+
print("missing ", resname, atName)
|
| 287 |
+
|
| 288 |
+
channels += [torch.tensor(haTMP, dtype=torch.float, device=device)]
|
| 289 |
+
channels = torch.torch.nn.utils.rnn.pad_sequence(channels, batch_first=True, padding_value=PADDING_INDEX)
|
| 290 |
+
return channels
|
| 291 |
+
|
| 292 |
+
|
| 293 |
+
def atomlistToRadius(atomList, hashing="FoldX_radius", device="cpu"):
|
| 294 |
+
"""
|
| 295 |
+
function to get radius from atom names (obtained parsing the pdb files with the parsePDB function)
|
| 296 |
+
|
| 297 |
+
|
| 298 |
+
|
| 299 |
+
Parameters
|
| 300 |
+
----------
|
| 301 |
+
atomNames : list
|
| 302 |
+
atom names obtained parsing the pdb files with the parsePDB function
|
| 303 |
+
hashing : FoldX_radius or dict
|
| 304 |
+
"FoldX_radius" provides the radius used by the FoldX force field
|
| 305 |
+
|
| 306 |
+
Alternatively, if you are not happy with the foldX radius, you can build a dictionary of dictionaries that defines the radius of every atom type in the pdb.
|
| 307 |
+
The first dictionary has the residue tag (three letters amino acid code) as key (3 letters compound name for hetero atoms, as written in the PDB file)
|
| 308 |
+
every residue key is associated to a dictionary, which the atom tags (as written in the PDB files) as keys and the radius (float) as value
|
| 309 |
+
|
| 310 |
+
for example, you can define the radius as following:
|
| 311 |
+
{
|
| 312 |
+
'CYS': {'N': 1.45, 'O': 1.37, 'C': 1.7, 'SG': 1.7, 'CB': 1.7, 'CA': 1.7}, # radius for cysteine atoms
|
| 313 |
+
'GLY': {'N': 1.45, 'O': 1.37, 'C': 1.7, 'CA': 1.7}, # radius for glycine atoms
|
| 314 |
+
...
|
| 315 |
+
'GOL': {'O1':1.37,'O2':1.37,'O3':1.37,'C1':1.7,'C2':1.7,'C3':1.7}, # radius for glycerol atoms
|
| 316 |
+
...
|
| 317 |
+
}
|
| 318 |
+
|
| 319 |
+
The default radius are the ones defined in FoldX
|
| 320 |
+
|
| 321 |
+
Radius default dictionary can be found in sources/hashings.py
|
| 322 |
+
|
| 323 |
+
device : torch.device
|
| 324 |
+
The device on which the model should run. E.g. torch.device("cuda") or torch.device("cpu:0")
|
| 325 |
+
Returns
|
| 326 |
+
-------
|
| 327 |
+
coords : torch.Tensor
|
| 328 |
+
coordinates of the atoms in the pdb file(s). Shape ( batch, numberOfAtoms, 3)
|
| 329 |
+
|
| 330 |
+
radius : torch.tensor
|
| 331 |
+
The radius of every atom. Shape (batch,numberOfAtoms)
|
| 332 |
+
|
| 333 |
+
"""
|
| 334 |
+
if hashing == "FoldX_radius":
|
| 335 |
+
hashing = hashings.radius
|
| 336 |
+
hahsingSomgleAtom = hashings.radiusSingleAtom
|
| 337 |
+
else:
|
| 338 |
+
assert type(hashing) is dict
|
| 339 |
+
|
| 340 |
+
radius = []
|
| 341 |
+
for singleAtomList in atomList:
|
| 342 |
+
haTMP = []
|
| 343 |
+
for i in singleAtomList:
|
| 344 |
+
resname = i.split("_")[0]
|
| 345 |
+
atName = i.split("_")[2]
|
| 346 |
+
if resname in hashing and atName in hashing[resname]:
|
| 347 |
+
haTMP += [hashing[resname][atName]]
|
| 348 |
+
elif atName[0] in hahsingSomgleAtom:
|
| 349 |
+
haTMP += [hahsingSomgleAtom[atName[0]]]
|
| 350 |
+
else:
|
| 351 |
+
haTMP += [1.0]
|
| 352 |
+
print("missing ", resname, atName)
|
| 353 |
+
radius += [torch.tensor(haTMP, dtype=torch.float, device=device)]
|
| 354 |
+
radius = torch.torch.nn.utils.rnn.pad_sequence(radius, batch_first=True, padding_value=PADDING_INDEX)
|
| 355 |
+
return radius
|
| 356 |
+
|
| 357 |
+
|
| 358 |
+
'''
|
| 359 |
+
def write_pdb(batchedCoords, atomNames , name=None, output_folder="outpdb/"): #I need to add the chain id
|
| 360 |
+
|
| 361 |
+
if name is None:
|
| 362 |
+
name = range(len(batchedCoords))
|
| 363 |
+
|
| 364 |
+
for struct in range(len(name)):
|
| 365 |
+
f = open(output_folder + str(name[struct]) + ".pdb", "w")
|
| 366 |
+
|
| 367 |
+
coords=batchedCoords[struct].data.numpy()
|
| 368 |
+
atname=atomNames[struct]
|
| 369 |
+
for i in range(len(coords)):
|
| 370 |
+
|
| 371 |
+
rnName = atname[i].split("_")[0]#hashings.resi_hash_inverse[resi_list[i]]
|
| 372 |
+
atName = atname[i].split("_")[2]#hashings.atom_hash_inverse[resi_list[i]][atom_list[i]]
|
| 373 |
+
pos = atname[i].split("_")[1]
|
| 374 |
+
chain = "A"
|
| 375 |
+
|
| 376 |
+
num = " " * (5 - len(str(i))) + str(i)
|
| 377 |
+
a_name = atName + " " * (4 - len(atName))
|
| 378 |
+
numres = " " * (4 - len(str(pos))) + str(pos)
|
| 379 |
+
|
| 380 |
+
x = round(float(coords[i][0]), 3)
|
| 381 |
+
sx = str(x)
|
| 382 |
+
while len(sx.split(".")[1]) < 3:
|
| 383 |
+
sx += "0"
|
| 384 |
+
x = " " * (8 - len(sx)) + sx
|
| 385 |
+
|
| 386 |
+
y = round(float(coords[i][1]), 3)
|
| 387 |
+
sy = str(y)
|
| 388 |
+
while len(sy.split(".")[1]) < 3:
|
| 389 |
+
sy += "0"
|
| 390 |
+
y = " " * (8 - len(sy)) + sy
|
| 391 |
+
|
| 392 |
+
z = round(float(coords[i][2]), 3)
|
| 393 |
+
sz = str(z)
|
| 394 |
+
while len(sz.split(".")[1]) < 3:
|
| 395 |
+
sz += "0"
|
| 396 |
+
z = " " * (8 - len(sz)) + sz
|
| 397 |
+
chain = " " * (2 - len(chain)) + chain
|
| 398 |
+
|
| 399 |
+
if rnName !="HET":
|
| 400 |
+
f.write("ATOM " + num + " " + a_name + "" + rnName + chain + numres + " " + x + y + z + " 1.00 64.10 " + atName[0] + "\n")
|
| 401 |
+
else:
|
| 402 |
+
f.write("HETATM" + num + " " + a_name + "" + rnName + chain + numres + " " + x + y + z + " 1.00 64.10 " + atName[0] + "\n")
|
| 403 |
+
'''
|