-
Notifications
You must be signed in to change notification settings - Fork 1
/
PPIDFT.py
198 lines (160 loc) · 6.2 KB
/
PPIDFT.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
# -*- coding: utf-8 -*-
'''
# Programs for dentifying protein-protein interaction by Fourier transform
#
# Changchuan Yin, Ph.D.
# Dept. of Mathematics, Statistics and Computer Science
# University of Illinois at Chicago
# Chicago, IL 60607
# USA
#
# Email: cyin1@uic.edu, cyinbox@gmail.com
# Last updated on 02/01/2017
#
# Citation
# Yin, C. & Yau, Stephen S.-T. (2017).A Coevolution Analysis for Identifying Protein-Protein Interactions by Fourier Transform, PLOS ONE
#
#
'''
import numpy as np
from scipy.fftpack import fft
from scipy.spatial import distance
import math
import os.path
from Bio import SeqIO
savePath = './PPIData'
#==============================================================================
# Function to return DFT real and image coefficients of a protein sequence
# Input: seq=protein sequence
# outputs: vRI=vector DFT coefficients of the protein sequence; ps = power spectrum
# The hydrophobicity value is from:
# Kyte, J., & Doolittle, R. F. (1982). A simple method for displaying the hydropathic character of a protein.
# Journal of molecular biology, 157(1), 105-132.
#==============================================================================
def DFTHydrophobicity(seq):
AAHydrophobicity = {'A': 1.8, 'R': -4.5, 'N': -3.5,'D': -3.5,'C':2.5, 'Q': -3.5,'E':-3.5,\
'G':-0.4,'H': -3.2,'I':4.5,'L':3.8,'K': -3.9,'M': 1.9,'F':2.8,'P':-0.9,\
'S': -0.8,'T':-0.7,'W':-0.9,'Y':-1.3,'V':4.2,'U':0.0,'X':0.0} #X is for padding zeros
n=len(seq)
listAA=np.zeros(n) #define zeros list of length n
for i in range(0,n): # use index from 0 to n
aa=seq[i]
listAA[i]=AAHydrophobicity[aa]
fftAA=fft(listAA)
ps=np.abs(fftAA[0:n])
ps=pow(ps,2)
R=fftAA.real
I=fftAA.imag
vRI=[]
for i in range (0,n):
vRI.append(R[i])
for i in range(0,n):
vRI.append(I[i])
return [ps,vRI]
#==============================================================================
# Function to compute the distance of two protein sequences in Euclidean space dimension size M
# Inputs: seq1, seq2 = protein sequences; M = vector space length
# outputs: dist = Euclidean distance of two proteins
#==============================================================================
def DFTProteinDistanceSpace(seq1,seq2,M):
n1=len(seq1)
n2=len(seq2)
if n1<M:
n=M-n1
seq1 = ''.join((seq1,'X'*n))
if n2<M:
n=M-n2
seq2 = ''.join((seq2,'X'*n))
[ps1,vRI1]=DFTHydrophobicity(seq1)
[ps2,vRI2]=DFTHydrophobicity(seq2)
dist = distance.euclidean(vRI1,vRI2)
return dist
#==============================================================================
# Function to compute the distance of two protein sequences in Euclidean space dimension size M
# Inputs: seqList = list of one protein sequences
# outputs: distM = dissimlarity distance matrix of mutual distance among proteins in the list
# distV = dissimlarity vector the mutual distance among proteins in the list
#==============================================================================
def mutualDistanceDFTProteins(seqList):
n=len(seqList)
distV=[]
distM=np.zeros([n, n])
lenM=[]
for i in range(0,n):
lenM.append(len(seqList[i]))
M=max(lenM)
for i in range(0,n-1):
seqA= seqList[i]
for j in range (i+1,n):
seqB= seqList[j]
dist=DFTProteinDistanceSpace(seqA,seqB,M)
distV.append(dist)
distM[i,j]=dist
return [distM,distV]
#==============================================================================
# Function to compute the DFT of a list of protein sequences of dimension size M
# Inputs: seqList = list of one protein sequences
# outputs: dftVectors = list of the DFT vectors for all the protein sequences in the list
#==============================================================================
def DFTProteinList(seqList):
n=len(seqList)
lenM=[]
for i in range(0,n):
lenM.append(len(seqList[i]))
M=max(lenM)
dftVectors=[]
seqP=''
for i in range(0,n):
seq= seqList[i]
s=len(seq)
p=M-s
seqP = ''.join((seq,'X'*p))
[ps,vRI]=DFTHydrophobicity(seqP)
dftVectors.append(vRI)
return dftVectors
#==============================================================================
# Function to remove zeros from two lists if zeros are in the same positions.
# Inputs: distA,distB = two lists of distances
# Output: distA, distB = two lists that were removed zeros if these zeros are in the same positions
#==============================================================================
def removeZerosSamePositions(distA,distB):
n=len(distA)
indexes=[]
for i in range (0,n):
if distA[i]==0 and distB[i]==0:
indexes.append(i)
for offset, index in enumerate(indexes):
index -= offset
del distA[index]
del distB[index]
return [distA, distB]
#==============================================================================
# Function to compute the score of PPI of two proteins
# Inputs: seqListA, seqListB= list of proteinA sequences, list of proteinA sequences;
# output: p = Peserson coefficient of the distance vectors of two proteins
#==============================================================================
def scorePPITreesP(seqListA,seqListB):
[distMA,distVA]=mutualDistanceDFTProteins(seqListA)
[distMA,distVB]=mutualDistanceDFTProteins(seqListB)
[distVA, distVB]=removeZerosSamePositions(distVA,distVB)
p=np.corrcoef(distVA, distVB)[0, 1] #Pearson coefficient
if math.isnan(p):
p=0
return p
#------------------------------------------------------------------------------
# Function to retrieve a specific gene/sequence from a fasta file
# Input: geneName=gene name,fastaFile=fasta file
# outputs: name,sequence
#------------------------------------------------------------------------------
def getAllSequences(proteinName):
fastaFile=proteinName+'.fasta'
fastaFile = os.path.join(savePath, fastaFile)
fasta_sequences = SeqIO.parse(open(fastaFile),'fasta')
name=''
sequence=''
listSeq=[]
for fasta in fasta_sequences:
[name, sequence] = fasta.id, str(fasta.seq)
listSeq.append(sequence)
#print('List',listSeq)
return listSeq