def write_pdb(balls, fn, strandlens, extraballs=False, reindexmap=None, mapfn=None):
tmpballs = []
pos = 0
for i in range(len(strandlens)):
strandlen = strandlens[i]
tmpstrandballs = []
for j in range(strandlen):
tmpstrandballs.append(balls[pos])
pos += 1
# remove the first and the last balls from the strands, which are extra balls
if not extraballs:
tmpstrandballs.pop(0)
tmpstrandballs.pop(0)
tmpstrandballs.pop(0)
tmpstrandballs.pop(0)
tmpstrandballs.pop(-1)
tmpstrandballs.pop(-1)
tmpstrandballs.pop(-1)
tmpstrandballs.pop(-1)
if i%2==1:
tmpstrandballs = tmpstrandballs[::-1]
tmpballs+=tmpstrandballs
chain = Bio.PDB.Chain.Chain('A')
for i in range(len(tmpballs)):
try:
res_id = (' ', tmpballs[i][Ball.p_resseqid], ' ')
restype = AA.index_to_three(tmpballs[i][Ball.p_aaid])
residue = Bio.PDB.Residue.Residue(res_id, restype, ' ')
cacoord = tmpballs[i].getcoord()
atom = Bio.PDB.Atom.Atom('CA', cacoord, 0, 0, ' ', 'CA', tmpballs[i][Ball.p_resseqid], 'C')
residue.add(atom)
chain.add(residue)
except:
res_id = ('A', tmpballs[i][Ball.p_resseqid], ' ')
restype = AA.index_to_three(tmpballs[i][Ball.p_aaid])
residue = Bio.PDB.Residue.Residue(res_id, restype, ' ')
cacoord = tmpballs[i].getcoord()
atom = Bio.PDB.Atom.Atom('CA', cacoord, 0, 0, ' ', 'CA', tmpballs[i][Ball.p_resseqid], 'C')
residue.add(atom)
chain.add(residue)
model = Bio.PDB.Model.Model(1)
model.add(chain)
structure = Bio.PDB.Structure.Structure("ref")
structure.add(model)
io = Bio.PDB.PDBIO()
io.set_structure(structure)
io.save(fn, write_end=False)
if reindexmap is not None and mapfn is not None:
np.savetxt(mapfn, reindexmap, fmt='%d')
def __init__(self, gc_file):
self.gc_file = gc_file
self.codons = dict()
self.amino_acids = dict()
self.has_CAI = False
with open(self.gc_file) as f:
for row in f:
l = row.strip("\n").split("\t")
self.amino_acids[l[0]] = AminoAcid(*l)
for c in l[3].split(","):
self.codons[c] = l[0]
def calculate_Energy(df, matrix):
radiusDict = LoadRadius()
CurrentAANitrogen = None
CurrentAACA = None
Currentresidue_num = None
EachAA = []
CurrentAA = None
for line in df.readlines():
if (line[0:4] != "ATOM"):
continue
element_list = extract_Data(line)
record_name = element_list[0]
atom_name = element_list[2]
residue_name = element_list[4]
alternate_indicator = element_list[3]
residue_num = element_list[-4]
xcor = float(element_list[-3])
ycor = float(element_list[-2])
zcor = float(element_list[-1])
if (atom_name == "H"):
continue
if (residue_name not in matrix):
continue
if (CurrentAA == None):
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
#If another amino acid begins
if (residue_num != Currentresidue_num):
state = CurrentAA.CalculateCenter()
if (state == False):
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
continue
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
EachAA.append(CurrentAA)
del CurrentAA
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
#If still the same amino acid
else:
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
state = CurrentAA.CalculateCenter()
if (state != False):
CurrentAA.CalculateCenter()
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
EachAA.append(CurrentAA)
#Scan over. Each amino acid is stored as an object in EachAA. Next step is to calculate the energy, results will be saved in EnergyList.
#Store the energy
E = 0
for m in range(len(EachAA)):
EachAA[m].EstablishCoordinate()
for n in range(len(EachAA)):
if (m == n):
continue
else:
dis = EachAA[m].DistanceBetweenAA(EachAA[n].center)
radiusSum = radiusDict[EachAA[m].name] + radiusDict[
EachAA[n].name]
if (
dis <= radiusSum
): #If the distance between two amino acid less than 10, we believe the two amino acid have interaction
rho, theta, phi = EachAA[m].ChangeCoordinate(
EachAA[n].center)
theta = min(int(math.floor(theta * 20 / np.pi)), 19)
phi = min(int(math.floor(phi * 10 / np.pi) + 10), 19)
E += matrix[EachAA[m].name][
EachAA[n].name][theta][phi] / rho
return E
def calculate_Energy(df,matrix):
# define some useful paramter
radiusDict = LoadRadius()
CurrentAANitrogen = None
CurrentAACA = None
Currentresidue_num = None
CurrentAA = None
# list of amino acids which have side chain
UseAA_list = []
# list of amino acids which have no side chain
IgnoreAA_list = []
# scan pdb file line one by one
for line in df.readlines():
if(line[0:4] != "ATOM"):
continue
# obtain information
element_list = extract_Data(line)
record_name = element_list[0]
atom_name = element_list[2]
residue_name = element_list[4]
alternate_indicator = element_list[3]
residue_num = element_list[-4]
xcor = float(element_list[-3])
ycor = float(element_list[-2])
zcor = float(element_list[-1])
# ignore hydrogen
if(atom_name == "H"):
continue
# ignore amino acid out of the list
if(residue_name not in matrix):
continue
# from here start to scan useful amino acid
# first amino acid
if(CurrentAA is None):
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if(atom_name == "N" or atom_name == "CA"):
if(alternate_indicator == " " or alternate_indicator == "A"):
if(atom_name == "N"):
CurrentAA.InputN(np.array([xcor,ycor,zcor]))
else:
CurrentAA.InputCA(np.array([xcor,ycor,zcor]))
else:
continue
if(residue_name == "GLY" or atom_name not in {"N","CA","C","O","O1","02"}):
if(alternate_indicator == " " or alternate_indicator == "A"):
CurrentAA.SumCenters(xcor,ycor,zcor,atom_name)
else:
continue
# current amino acid is not the first
else:
#If another amino acid begins
if(residue_num != Currentresidue_num):
state = CurrentAA.Check()
# previous amino acid has no problem
if(state == True):
CurrentAA.CalculateCenter()
UseAA_list.append(CurrentAA)
# previous amino acid has problem
else:
info = [state,Currentresidue_num]
IgnoreAA_list.append(info)
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if(atom_name == "N" or atom_name == "CA"):
if(alternate_indicator == " " or alternate_indicator == "A"):
if(atom_name == "N"):
CurrentAA.InputN(np.array([xcor,ycor,zcor]))
else:
CurrentAA.InputCA(np.array([xcor,ycor,zcor]))
else:
continue
if(residue_name == "GLY" or atom_name not in {"N","CA","C","O","O1","02"}):
if(alternate_indicator == " " or alternate_indicator == "A"):
CurrentAA.SumCenters(xcor,ycor,zcor,atom_name)
else:
continue
#If still the same amino acid
else:
if(atom_name == "N" or atom_name == "CA"):
if(alternate_indicator == " " or alternate_indicator == "A"):
if(atom_name == "N"):
CurrentAA.InputN(np.array([xcor,ycor,zcor]))
else:
CurrentAA.InputCA(np.array([xcor,ycor,zcor]))
else:
continue
if(residue_name == "GLY" or atom_name not in {"N","CA","C","O","O1","02"}):
if(alternate_indicator == " " or alternate_indicator == "A"):
CurrentAA.SumCenters(xcor,ycor,zcor,atom_name)
else:
continue
state = CurrentAA.Check()
if(state == True):
CurrentAA.CalculateCenter()
UseAA_list.append(CurrentAA)
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
else:
info = [state,Currentresidue_num]
IgnoreAA_list.append(info)
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
# Scan over. Each amino acid is stored as an object in UseAA_list.
# Next step is to calculate the energy, results will be saved in EnergyList.
E = 0
for m in range(len(UseAA_list)):
UseAA_list[m].EstablishCoordinate()
for n in range(len(UseAA_list)):
if(m == n):
continue
else:
dis = UseAA_list[m].DistanceBetweenAA(UseAA_list[n].center)
radiusSum = radiusDict[UseAA_list[m].name] + radiusDict[UseAA_list[n].name]
if(dis <= radiusSum):#If the distance between two amino acid less than 10, we believe the two amino acid have interaction
rho,theta,phi = UseAA_list[m].ChangeCoordinate(UseAA_list[n].center)
theta = min(int(math.floor(theta*20/np.pi)),19)
phi = min(int(math.floor(phi*10/np.pi) + 10),19)
E += matrix[UseAA_list[m].name][UseAA_list[n].name][theta][phi] / rho
return E
def _construct_ideal_balls_(self):
# load residues
residues = np.loadtxt('inputs/{pdb}/{pdb}.res'.format(pdb=self.pdb)).astype(int).tolist()
residues = [200]+residues
# load regs and peris got from reg adjustment
periregs = np.loadtxt('{inputdirn}/regs/{pdb}.regs'.format(inputdirn=self.inputdirn, pdb=self.pdb)).astype(int)
peris = np.loadtxt('{inputdirn}/peris/{pdb}.peris'.format(inputdirn=self.inputdirn, pdb=self.pdb)).astype(int)
# load facings got from reg adjustment
with open('{inputdirn}/facings/{pdb}.facings'.format(inputdirn=self.inputdirn, pdb=self.pdb)) as f:
lines = f.readlines()
firstfacings = [ line.strip() for line in lines ]
# load strandends
strandends = np.loadtxt('inputs/{pdb}/{pdb}.strands'.format(pdb=self.pdb)).astype(int)
# the strand ranges here may not be the ranges used for reg prediction
# correct reg according to the difference between strandends to construct barrel and peris used in reg pred
# after this loop, array periregs will be the predicted regs of strandends
for strdi in range(len(strandends)):
strdim1 = (strdi-1)%len(strandends)
if strdi%2==0:
periregs[strdi] -= strandends[strdi][0]-peris[strdi]
periregs[strdim1] += strandends[strdi][0]-peris[strdi]
else:
periregs[strdi] += strandends[strdi][1]-peris[strdi]
periregs[strdim1] -= strandends[strdi][1]-peris[strdi]
# correct facing according to the difference between strandends to construct barrel and peris used in reg pred
for strdi in range(len(strandends)):
if strdi%2==0:
if (strandends[strdi][0]-peris[strdi])%2!=0:
if firstfacings[strdi]=='OUT':
firstfacings[strdi]='IN'
else:
firstfacings[strdi]='OUT'
else:
if (strandends[strdi][1]-peris[strdi])%2!=0:
if firstfacings[strdi]=='OUT':
firstfacings[strdi]='IN'
else:
firstfacings[strdi]='OUT'
# add extra residues for bbq
for strdi in range(len(strandends)):
strandends[strdi][0]-=Barrel.extra_ball_num
strandends[strdi][1]+=Barrel.extra_ball_num
# construct facing arrays for all residues (including extra residues)
facings = []
for fac in firstfacings:
if fac == 'OUT':
facings.append([Ball.Facing.OUT])
else:
facings.append([Ball.Facing.IN])
for strdi in range(len(strandends)):
for resi in range(strandends[strdi][1]-strandends[strdi][0]):
if facings[strdi][resi] == Ball.Facing.OUT:
facings[strdi].append(Ball.Facing.IN)
else:
facings[strdi].append(Ball.Facing.OUT)
peripositions = np.cumsum( np.hstack( ([0], -periregs) ) )
N = len(strandends) # strand num
A = self.A # intrastrand Ca distance
B = self.B # interstrand Ca distance
S = sum(periregs) # shear number
## circle formula
#a = math.sqrt( (N*B)**2+(S*A)**2 ) / 2.0 / math.pi # tilt angle
#theta = math.asin(S*A/2.0/math.pi/a) # radius
## polygan formula
theta = math.atan( S*A / (N*B) ) # tilt angle
a = B / ( 2*math.sin(math.pi/N) * math.cos(theta) ) # radius
self.radius = a
b = a / math.tan(theta) # vertical speed
c = math.sqrt( a*a + b*b )
delta = 2 * math.pi * a * a / (c*N) # offset on the neigbouring strand to ensure inter H-bond is perpendicular to the strand
currid = 0
ids = []
seqids = []
restypes = []
cacoords = []
# construct the barrel
for strdi in range(N):
# seq ids
if strdi%2==0:
seqids.append( range( strandends[strdi][0], strandends[strdi][1]+1 ) )
else:
seqids.append( range( strandends[strdi][0], strandends[strdi][1]+1 )[::-1] )
ids.append( range( currid, currid+len(seqids[strdi]) ) )
currid += len(seqids[strdi])
# residue types
restypes.append([])
for seqid in seqids[strdi]:
try:
restypes[strdi].append(AA.index_to_one(residues[seqid]))
except:
restypes[strdi].append('C')
cacoords.append([])
for resi in range(strandends[strdi][1]-strandends[strdi][0]+1):
# zigzag deviation
if facings[strdi][resi] == Ball.Facing.OUT:
dr = self.dr
# righthand side of out facing residue is always SH
# lefthand side NH
if strdi%2==0:
dw = self.dw
else:
dw = -self.dw
if self.np1_right: #test TODO
if strdi%2==0:
dw = -self.dw
else:
dw = self.dw
else:
dr = -self.dr
if strdi%2==0:
dw = -self.dw
else:
dw = self.dw
if self.np1_right: #test TODO
if strdi%2==0:
dw = self.dw
else:
dw = -self.dw
s = ( peripositions[strdi] + resi ) * A + strdi * delta
x = (a+dr) * math.cos(s/c-2*math.pi*strdi/N);
y = (a+dr) * math.sin(s/c-2*math.pi*strdi/N);
if self.np1_right: #test TODO
s = ( peripositions[strdi] + resi ) * A + (N-strdi) * delta
x = (a+dr) * math.sin(s/c+2*math.pi*strdi/N);
y = (a+dr) * math.cos(s/c+2*math.pi*strdi/N);
z = b * s/c;
xn1 = (a+dr) * ( - math.cos(s/c-2*math.pi*strdi/N) + math.cos((s+delta)/c-2*math.pi*(strdi+1)/N) );
yn1 = (a+dr) * ( - math.sin(s/c-2*math.pi*strdi/N) + math.sin((s+delta)/c-2*math.pi*(strdi+1)/N) );
zn1 = b*delta/c
if (strdi%2==1 and facings[strdi][resi] == Ball.Facing.OUT) or (strdi%2==0 and facings[strdi][resi] == Ball.Facing.IN):
xn1 = (a+dr) * ( - math.cos(s/c-2*math.pi*(strdi-1)/N) + math.cos((s+delta)/c-2*math.pi*strdi/N) );
yn1 = (a+dr) * ( - math.sin(s/c-2*math.pi*(strdi-1)/N) + math.sin((s+delta)/c-2*math.pi*strdi/N) );
zn1 = b*delta/c
if self.np1_right: #test TODO
xn1 = (a+dr) * ( - math.sin(s/c+2*math.pi*strdi/N) + math.sin((s+delta)/c+2*math.pi*(strdi+1)/N) );
yn1 = (a+dr) * ( - math.cos(s/c+2*math.pi*strdi/N) + math.cos((s+delta)/c+2*math.pi*(strdi+1)/N) );
if (strdi%2==1 and facings[strdi][resi] == Ball.Facing.OUT) or (strdi%2==0 and facings[strdi][resi] == Ball.Facing.IN):
xn1 = (a+dr) * ( - math.sin(s/c+2*math.pi*(strdi-1)/N) + math.sin((s+delta)/c+2*math.pi*strdi/N) );
yn1 = (a+dr) * ( - math.cos(s/c+2*math.pi*(strdi-1)/N) + math.cos((s+delta)/c+2*math.pi*strdi/N) );
n1norm = math.sqrt(xn1*xn1+yn1*yn1+zn1*zn1)
xn1 = xn1/n1norm
yn1 = yn1/n1norm
zn1 = zn1/n1norm
x+=dw*xn1
y+=dw*yn1
z+=dw*zn1
cacoords[strdi].append(np.array([x,y,z]))
self.strandlens.append(len(ids[strdi]))
for i in range(len(ids)):
for j in range(len(ids[i])):
## following line is for model/param selections
#ball = Ball([ ids[i][j], seqids[i][j], cacoords[i][j][0], cacoords[i][j][1], cacoords[i][j][2], AA.one_to_index(restypes[i][j]), facings[i][j] ])
## store ballids instead of seqids. needs to be correted after bbq
## if using seqids, bbq will have problems
if i%2!=0:
ball = Ball([ ids[i][j], ids[i][len(ids[i])-j-1], cacoords[i][j][0], cacoords[i][j][1], cacoords[i][j][2], AA.one_to_index(restypes[i][j]), facings[i][j] ])
if j >= Barrel.extra_ball_num and j < len(ids[i])-Barrel.extra_ball_num:
self.reindexmap.append( (ids[i][len(ids[i])-j-1], seqids[i][j]) )
else:
ball = Ball([ ids[i][j], ids[i][j], cacoords[i][j][0], cacoords[i][j][1], cacoords[i][j][2], AA.one_to_index(restypes[i][j]), facings[i][j] ])
if j >= Barrel.extra_ball_num and j < len(ids[i])-Barrel.extra_ball_num:
self.reindexmap.append( (ids[i][j], seqids[i][j]) )
self.balls.append(ball)
def __str__(self): return str(self[Ball.p_ballid]) +" "+ str(self[Ball.p_resseqid]) +" " +\ str(self.getcoord())+" "+\ AA.index_to_one(self[Ball.p_aaid]) +" "+ str(self[Ball.p_facing])
def processAAforchian(chain,aaDict):
CurrentAANitrogen = None
CurrentAACA = None
Currentresidue_num = None
EachAA = []
CurrentAA = None
for line in chain:
if (line[0:4] != "ATOM"):
continue
element_list = extract_Data(line)
record_name = element_list[0]
atom_name = element_list[2]
residue_name = element_list[4]
alternate_indicator = element_list[3]
residue_num = element_list[-4]
chain_id = element_list[-5]
xcor = float(element_list[-3])
ycor = float(element_list[-2])
zcor = float(element_list[-1])
if (atom_name == "H"):
continue
if (residue_name not in aaDict):
continue
if (CurrentAA == None):
CurrentAA = AA.AminoAcid(residue_name, residue_num, chain_id)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY" or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
# If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
# If another amino acid begins
if (residue_num != Currentresidue_num):
state = CurrentAA.CalculateCenter()
if (state == False):
CurrentAA = AA.AminoAcid(residue_name, residue_num, chain_id)
Currentresidue_num = residue_num
continue
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
CurrentAA.EstablishCoordinate()
# Amino Acid check
EachAA.append(CurrentAA)
del CurrentAA
CurrentAA = AA.AminoAcid(residue_name, residue_num, chain_id)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY" or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
# If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
# If still the same amino acid
else:
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY" or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
# If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
state = CurrentAA.CalculateCenter()
if (state != False):
#CurrentAA.CalculateCenter()
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
CurrentAA.EstablishCoordinate()
EachAA.append(CurrentAA)
return EachAA
def process_AA2(AA_information, matrix):
CurrentAA = None
CurrentAANitrogen = None
CurrentAACA = None
Currentresidue_num = None
#for debug
lines = AA_information
for line in lines:
if (line[0:4] != "ATOM"):
continue
element_list = extract_Data(line)
record_name = element_list[0]
atom_name = element_list[2]
residue_name = element_list[4]
alternate_indicator = element_list[3]
#do some change
residue_num = element_list[-4]
#add chain_id
chain_id = element_list[-5]
xcor = float(element_list[-3])
ycor = float(element_list[-2])
zcor = float(element_list[-1])
if (atom_name == "H"):
continue
if (residue_name not in matrix):
continue
if (CurrentAA == None):
CurrentAA = AA.AminoAcid(residue_name, residue_num, chain_id)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
#if(alternate_indicator == "A" and line[15] == "1"):
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
#If another amino acid begins
if (residue_num != Currentresidue_num):
state = CurrentAA.CalculateCenter()
if (state == False):
CurrentAA = AA.AminoAcid(residue_name, residue_num,
chain_id)
Currentresidue_num = residue_num
#continue
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
#residue_name='ALA'
#all_amino_acids.append(CurrentAA)
del CurrentAA
CurrentAA = AA.AminoAcid(residue_name, residue_num, chain_id)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
#if(alternate_indicator == "A" and line[15] == "1"):
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
#If still the same amino acid
else:
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
#if(alternate_indicator == "A" and line[15] == "1"):
if (alternate_indicator == "A"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
CurrentAA.CalculateCenter()
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
#CurrentAA.EstablishCoordinate()
return CurrentAA
def main(protein_fasta_open_file, list_codon_usage_open_files, output_destination, restriction_enzymes=""):
# parse protein
record = Parser.parse_fasta_file(protein_fasta_open_file)
name, id, sequence = record.name, record.id, record.seq
creatures = {}
# parse table
if len(list_codon_usage_open_files) == 0:
raise Exception("Error: Empty codon table filnames")
# parses organism files , assuming they are already open
for fname, open_file in list_codon_usage_open_files:
creature_name = fname.split('.')[0]
codon_usage_dict, codon_to_protein_dict, AA_list = Parser.parse_kazusa_codon_usage_table(open_file)
creatures[creature_name] = codon_usage_dict, codon_to_protein_dict, AA_list
# creates AA
Amino_Acids_obj_list = []
AA_LIST = creatures[creature_name][2]
codon_to_protein_dict = creatures[creature_name][1]
for aa in AA_LIST:
AA = AminoAcid.AminoAcid(aa, codon_to_protein_dict)
Amino_Acids_obj_list.append(AA)
for creature_name, creature_tuple in creatures.items():
codon_usage_dict, codon_to_protein_dict, AA_list = creature_tuple
for AA in Amino_Acids_obj_list:
AA.add_organism_codons(codon_usage_dict, creature_name)
prot_analisys = ProtParam.ProteinAnalysis(sequence._data)
aa_count_dict = prot_analisys.count_amino_acids()
# replaces aa with codons from codon pool
ouput_protein_list = Calculator.compute_and_Switch(Amino_Acids_obj_list, sequence, aa_count_dict)
final_sequence = "".join(ouput_protein_list)
final_sequence = final_sequence.replace("U", "T")
# analyse final sequance
if len(final_sequence) != len(sequence) * 3:
raise Exception("final sequance length does not match input sequence length")
# output_file_name = os.path.join(output_destination, "Ouput.fasta")
record = SeqRecord.SeqRecord(Seq(final_sequence, ), name=name)
if record.translate().seq != sequence:
raise Exception("error- resulting DNA does not translate back to protein")
# restriction enzymes- verifies they do not cut the sequence. if they do, pick the least cut sequence
if restriction_enzymes != "":
restriction_enzymes_list = restriction_enzymes.replace(",", " ").replace('\n', ' ').replace("\t", " ").split()
batch = RestrictionBatch(restriction_enzymes_list)
num_cutting = len(check_restriction(Seq(final_sequence, generic_dna), batch))
best_num_cutting = np.inf
best_sequ = final_sequence
iterations = 100
no_enzymes_cut = num_cutting == 0
# if the original sequence had a restriction site, repeat the sequence building 100 times , or until
# a non- cut sequence is found
while iterations > 0 and num_cutting > 0:
ouput_protein_list = Calculator.compute_and_Switch(Amino_Acids_obj_list, sequence, aa_count_dict)
final_sequence = "".join(ouput_protein_list)
final_sequence = final_sequence.replace("U", "T")
# analyse final sequance
if len(final_sequence) != len(sequence) * 3:
raise Exception("final sequance length does not match input sequence length")
# output_file_name = os.path.join(output_destination, "Ouput.fasta")
record = SeqRecord.SeqRecord(Seq(final_sequence, generic_dna), name=name)
if record.translate().seq != sequence:
print("error- resulting DNA does not translate back to protein")
exit(1)
# if achieved non cutting sequence, save and return
num_cutting = len(check_restriction(Seq(final_sequence, generic_dna), batch))
if num_cutting == 0:
check_restriction(Seq(final_sequence, generic_dna), batch, to_print=True)
print("printing to output file....")
SeqIO.write(record, output_destination, "fasta")
print("ouput sucsessful")
return "Output Sucsessful"
best_num_cutting = min(best_num_cutting, num_cutting)
if best_num_cutting == num_cutting:
best_sequ = final_sequence
iterations -= 1
# return best sequence, as in one that is cut by the least amount of restriction enzymes
if best_num_cutting > 0:
cutting = check_restriction(Seq(best_sequ, generic_dna), batch, to_print=True)
record = SeqRecord.SeqRecord(Seq(best_sequ, generic_dna), name=name)
SeqIO.write(record, output_destination, "fasta")
return "The enzymes the cut the sequence are:" + str(cutting) + "\n Output printed to specified location."
SeqIO.write(record, output_destination, "fasta")
return "ouput sucsessful"
def ProcessPDB(chainlines, matrix):
#df = open(file,'r')
radiusDict = LoadRadius()
CurrentAANitrogen = None
CurrentAACA = None
Currentresidue_num = None
EachAA = []
CurrentAA = None
for line in chainlines:
if (line[0:4] != "ATOM"):
continue
element_list = extract_Data(line)
record_name = element_list[0]
atom_name = element_list[2]
residue_name = element_list[4]
alternate_indicator = element_list[3]
residue_num = element_list[-4]
xcor = float(element_list[-3])
ycor = float(element_list[-2])
zcor = float(element_list[-1])
if (atom_name == "H"):
continue
if (residue_name not in matrix):
continue
if (CurrentAA == None):
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
#If another amino acid begins
if (residue_num != Currentresidue_num):
state = CurrentAA.CalculateCenter()
if (state == False):
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
continue
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
EachAA.append(CurrentAA)
del CurrentAA
CurrentAA = AA.AminoAcid(residue_name)
Currentresidue_num = residue_num
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
#If still the same amino acid
else:
if (atom_name == "N" or atom_name == "CA"):
if (alternate_indicator == "B"):
continue
if (atom_name == "N"):
CurrentAANitrogen = np.array([xcor, ycor, zcor])
else:
CurrentAACA = np.array([xcor, ycor, zcor])
if (residue_name == "GLY"
or atom_name not in {"N", "CA", "C", "O", "O1", "02"}):
if (alternate_indicator != " "):
#If cases like "AASN or BASN" appears, we only add A
if (alternate_indicator == "A" and line[15] == "1"):
CurrentAA.SumCenters(xcor, ycor, zcor)
else:
continue
else:
CurrentAA.SumCenters(xcor, ycor, zcor)
state = CurrentAA.CalculateCenter()
if (state != False):
CurrentAA.CalculateCenter()
CurrentAA.InputCAN(CurrentAANitrogen, CurrentAACA)
EachAA.append(CurrentAA)
return EachAA
def main( fn, fastafile ): genetic_code = dict() codon_dict = dict() with open( fn ) as f: for row in f: l = row.strip( "\n" ).split( "\t" ) genetic_code[l[0]] = AminoAcid( *l ) for c in l[3].split( "," ): c = c.replace( "U", "T" ) codon_dict[c] = l[0] #print genetic_code, len( genetic_code ) #print #print codon_dict, len( codon_dict ) #print #for a in genetic_code: #print genetic_code[a] #print orf_frame = dict() with open( "/Users/paulkorir/Dropbox/Euplotes/FrameshiftPredictionData/one_orfs.txt" ) as f: for row in f: if row[0] == "T": continue l = row.strip( "\n" ).split( "\t" ) orf_frame[l[0]] = ORFInfo( *l ) # read in the data from the fasta file total = 0 ok_count = 0 nok_count = 0 for seq_record in SeqIO.parse( fastafile, 'fasta' ): sequence = str( seq_record.seq ) seq_name = seq_record.id.split( " " )[0] # get the position of the first ATG frame = orf_frame[seq_name].frame last = orf_frame[seq_name].last i = frame start = None while i <= len( sequence ) - 3: codon = sequence[i:i+3] if codon == "ATG": start = i break else: i += 3 if start == None: print >> sys.stderr, "Missing ATG in frame %d in sequence %s" % ( frame, seq_name ) total += 1 nok_count += 1 continue #if ( last - start + 1 ) % 3 == 0: #ok_count += 1 #total += 1 #else: #print seq_name, start, last, ( last - start ) + 1, ( last - start + 1 ) % 3, i #nok_count += 1 #total += 1 #print ok_count/total, nok_count/total #print ok_count, nok_count, total # make sure it's in the first coding frame cds = sequence[start:last+1] print ">" + seq_record.id print cds
def SingleStructure(decoyname,DecoyPath,model_path,radius_path):
cdDict={"ALA":{},"VAL":{},"LEU":{},"ILE":{},"PHE":{},\
"TRP":{},"MET":{},"PRO":{},"GLY":{},"SER":{},\
"THR":{},"CYS":{},"TYR":{},"ASN":{},"GLN":{},\
"HIS":{},"LYS":{},"ARG":{},"ASP":{},"GLU":{},}
cdDict = loadModel(model_path,cdDict)
radiusDict = LoadRadius(radius_path)
if(decoyname == "native.pdb"):
xcor = 6
ycor = 7
zcor = 8
AAnum = 5
else:
xcor = 5
ycor = 6
zcor = 7
AAnum = 4
df = open(DecoyPath)
#CurrentAAName = None
CurrentAANitrogen = None
CurrentAACA = None
CurrentAANumber = None
EachAA = []
CurrentAA = None
for line in df.readlines():
#print line
Element,AAtype,AANUMBER = ExtractData(line)
if(Element[0] != "ATOM"):
CurrentAA.CalculateCenter()
CurrentAA.InputCAN(CurrentAANitrogen,CurrentAACA)
EachAA.append(CurrentAA)
continue
if(Element[2] == "H"):
continue
if(AAtype not in cdDict):
continue
if(CurrentAA == None):
#print("First object establised")
#CurrentAAName = Element[3]
CurrentAA = AA.AminoAcid(AAtype)
CurrentAANumber = AANUMBER
if(Element[2] == "N" or Element[2] == "CA"):
if(line[16] == "B"):
continue
if(Element[2] == "N"):
CurrentAANitrogen = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
else:
CurrentAACA = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
if(AAtype == "GLY" or Element[2] not in {"N","CA","C","O","O1","02"}):
if(line[16] != " "):
#If cases like "AASN or BASN" appears, we only add A
if(line[16] == "A" and line[15] == "1"):
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
else:
continue
else:
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
else:
#If another amino acid begins
if(AANUMBER != CurrentAANumber):
#print CurrentAA.AminoAcidAmount
#print CurrentAAName,Element[3]
state = CurrentAA.CalculateCenter()
if(state == False):
CurrentAA = AA.AminoAcid(AAtype)
CurrentAANumber = AANUMBER
continue
CurrentAA.InputCAN(CurrentAANitrogen,CurrentAACA)
#print sys.getrefcount(CurrentAA)
EachAA.append(CurrentAA)
del CurrentAA
CurrentAA = AA.AminoAcid(AAtype)
#print sys.getrefcount(CurrentAA)
CurrentAANumber = AANUMBER
if(Element[2] == "N" or Element[2] == "CA"):
if(line[16] == "B"):
continue
if(Element[2] == "N"):
CurrentAANitrogen = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
else:
CurrentAACA = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
if(AAtype == "GLY" or Element[2] not in {"N","CA","C","O","O1","02"}):
if(line[16] != " "):
#If cases like "AASN or BASN" appears, we only add A
if(line[16] == "A" and line[15] == "1"):
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
else:
continue
else:
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
#If still the same amino acid
else:
if(Element[2] == "N" or Element[2] == "CA"):
if(line[16] == "B"):
continue
if(Element[2] == "N"):
CurrentAANitrogen = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
else:
CurrentAACA = np.array([float(Element[xcor]),float(Element[ycor]),float(Element[zcor])])
if(AAtype == "GLY" or Element[2] not in {"N","CA","C","O","O1","02"}):
if(line[16] != " "):
#If cases like "AASN or BASN" appears, we only add A
if(line[16] == "A" and line[15] == "1"):
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
else:
continue
else:
CurrentAA.SumCenters(float(Element[xcor]),float(Element[ycor]),float(Element[zcor]))
del CurrentAA#Free the current object.
#Scan over. Each amino acid is stored as an object in EachAA. Next step is to calculate the energy, results will be saved in EnergyList.
E = 0 #Store the energy
Time = 0
for m in range(len(EachAA)):
#Establish axis first
EachAA[m].EstablishCoordinate()
for n in range(len(EachAA)):
if(m == n):
continue
else:
dis = EachAA[m].DistanceBetweenAA(EachAA[n].center)
radiusSum = radiusDict[EachAA[m].name] + radiusDict[EachAA[n].name]
if(dis <= radiusSum):#If the distance between two amino acid less than 10, we believe the two amino acid have interaction
#print EachAA[m].ChangeCoordinate(EachAA[n].center)
rho,theta,phi = EachAA[m].ChangeCoordinate(EachAA[n].center)
theta = min(int(math.floor(theta*20/np.pi)),19)
phi = min(int(math.floor(phi*10/np.pi) + 10),19)
#print EachAA[m].name,EachAA[n].name
E += cdDict[EachAA[m].name][EachAA[n].name][theta][phi] / rho
Time += 1
return E,Time
def main(protein_fasta_filename, list_codon_usage_filenames,output_destination, restriction_enzymes="" ):
#verify input
verify_input()
#parse protein
record= Parser.parse_fasta_file(protein_fasta_filename)
name, id, sequence =record.name, record.id, record.seq
creatures = {}
#parse table
if len(list_codon_usage_filenames) ==0:
print("Error: Empty codon table filnames")
exit(1)
for i, file_name in enumerate(list_codon_usage_filenames):
creature_name = ntpath.basename(file_name).split('.')[0] #TODO watch out
codon_usage_dict, codon_to_protein_dict, AA_list = Parser.parse_kazusa_codon_usage_table(str(file_name))
creatures[creature_name] = codon_usage_dict, codon_to_protein_dict, AA_list
#creates AA
Amino_Acids_obj_list =[]
AA_LIST= creatures[creature_name][2]
codon_to_protein_dict = creatures[creature_name][1]
for aa in AA_LIST:
AA = AminoAcid.AminoAcid(aa,codon_to_protein_dict )
Amino_Acids_obj_list.append(AA)
for creature_name, creature_tuple in creatures.items():
codon_usage_dict, codon_to_protein_dict, AA_list = creature_tuple
for AA in Amino_Acids_obj_list:
AA.add_organism_codons(codon_usage_dict, creature_name)
prot_analisys = ProtParam.ProteinAnalysis(sequence._data)
aa_count_dict = prot_analisys.count_amino_acids()
ouput_protein_list = Calculator.compute_and_Switch(Amino_Acids_obj_list, sequence,aa_count_dict)
final_sequence = "".join(ouput_protein_list)
#analyse final sequance
if len(final_sequence) != len(sequence) * 3:
print("final sequance length does not match input sequence length")
exit(1)
output_file_name = os.path.join (output_destination ,"Ouput.fasta" )
record = SeqRecord.SeqRecord(Seq(final_sequence, generic_dna) , name = name )
if record.translate().seq != sequence:
print("error- resulting DNA does not translate back to protein")
exit(1)
#restriction enzymes
if restriction_enzymes != "":
restriction_enzymes_list = restriction_enzymes.replace(",", " ").replace('\n', ' ').replace("\t"," ").split()
batch = RestrictionBatch(restriction_enzymes_list)
num_cutting = check_restriction(Seq(final_sequence), batch)
iterations = 100
while iterations> 0 and num_cutting > 0 :
ouput_protein_list = Calculator.compute_and_Switch(Amino_Acids_obj_list, sequence, aa_count_dict)
final_sequence = "".join(ouput_protein_list)
# analyse final sequance
if len(final_sequence) != len(sequence) * 3:
print("final sequance length does not match input sequence length")
exit(1)
output_file_name = os.path.join(output_destination, "Ouput.fasta")
record = SeqRecord.SeqRecord(Seq(final_sequence, generic_dna), name=name)
if record.translate().seq != sequence:
print("error- resulting DNA does not translate back to protein")
exit(1)
num_cutting = check_restriction(Seq(final_sequence), batch)
iterations -= 1
print("printing to output file....")
with open(output_file_name, "w") as output_handle:
SeqIO.write(record, output_handle, "fasta")
print("ouput sucsessful")
return True