In [1]:

    

import sys
sys.path.append("/projects/pw8/wl45/open3spn2")
sys.path.append("/projects/pw8/wl45/openawsem")
sys.path.append("/Users/weilu/open3spn2")
sys.path.append("/Users/weilu/openmmawsem")
sys.path.append("C:/Users/luwei/Documents/GitHub/open3spn2")
sys.path.append("C:/Users/luwei/Documents/GitHub/openawsem")


import open3SPN2
import ffAWSEM
import time
import simtk.openmm
import simtk.openmm.app


    

In [10]:

    

import scipy.spatial.distance as sdist
import pandas as pd


    

In [3]:

    

pdb_file = "/Users/weilu/Research/server/jun_week1_2020/protein_DNA_benchmark/DNAProtein_Platform_OpenCL_date_20200226_pdb_1a36_repetition_0_clean.pdb"
seq_file = "/Users/weilu/Research/server/jun_week1_2020/protein_DNA_benchmark/DNAProtein_Platform_OpenCL_date_20200226_pdb_1a36_repetition_0_protein.seq"
with open(seq_file) as ps:
    protein_sequence_one=ps.readlines()[0]
protein=ffAWSEM.Protein.fromCoarsePDB(pdb_file,sequence=protein_sequence_one)


    

In [6]:

    

data = protein.atoms


    

In [27]:

    

protein_resNames = ["NGP", "IGL", "IPR", "NTER", "CTER"]
DNA_resNames = ["DA", "DC", "DT", "DG"]
group1_index = []
group2_index = []
for i, line in data.iterrows():
    resname = line["resname"]
    index = line["serial"]
    name = line["name"]
    if resname in DNA_resNames and name == "P":
        group1_index.append(index)
    if resname in protein_resNames and name == "CA":
        group2_index.append(index)


    

In [11]:

    

#Calculate native distances
CA_atoms=protein.atoms[protein.atoms.name=='CA']
P_atoms=protein.atoms[protein.atoms.name=='P']
d=sdist.cdist(CA_atoms[['x','y','z']],P_atoms[['x','y','z']])/10 #Distance in nanometers
d_sq=pd.DataFrame(d,index=CA_atoms.index,columns=P_atoms.index)
d=d_sq.copy()
d['Protein']=d.index
d=d.melt(id_vars=['Protein'])
d=d.rename(columns={'variable': 'DNA', 'value': 'distance'})


    

In [23]:

    

d.query("distance < 1")


    

    
Out[23]:







  

    

      

      
Protein
      
DNA
      
distance
    
  
  

    

      
352
      
2209
      
3
      
0.993812
    
    

      
353
      
2215
      
3
      
0.982950
    
    

      
354
      
2221
      
3
      
0.909546
    
    

      
356
      
2233
      
3
      
0.853868
    
    

      
357
      
2238
      
3
      
0.817565
    
    

      
...
      
...
      
...
      
...
    
    

      
20757
      
2315
      
116
      
0.692107
    
    

      
20758
      
2321
      
116
      
0.960002
    
    

      
20759
      
2327
      
116
      
0.996391
    
    

      
21132
      
1272
      
119
      
0.958682
    
    

      
21683
      
1272
      
122
      
0.901890
    
  

268 rows × 3 columns

In [17]:

    

d["DNA"].unique()


    

    
Out[17]:





array([3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, 39, 42, 45, 48, 51,
       54, 57, 60, 63, 68, 71, 74, 77, 80, 83, 86, 89, 92, 95, 98, 101,
       104, 107, 110, 113, 116, 119, 122, 125, 128], dtype=object)

In [18]:

    

d.shape


    

    
Out[18]:





(23142, 3)

In [19]:

    

23142/40


    

    
Out[19]:





578.55

In [31]:

    

data = pd.read_csv("/Users/weilu/Research/server/jun_week3_2020/protein_DNA/selected.csv", index_col=0)

sampled = data.sample(6, random_state=28)
pdb_list = sampled.idcode.to_list()


    

In [38]:

    




    

In [41]:

    

pdb_list = sampled.idcode.to_list()


    

In [42]:

    

pdb_list = [a.lower() for a in pdb_list]


    

In [43]:

    

print(pdb_list)


    

    




['4y60', '5ke8', '1a1j', '5lxu', '1skn', '6a2h']

In [35]:

    

data.sample(6, random_state=0)


    

    
Out[35]:







  

    

      

      
Number of missing residues
      
Resolution
      
idcode
      
release_date
      
First
      
Chain DNA1 ID
      
Chain DNA1 Compound
      
Chain DNA2 ID
      
Chain DNA2 Compound
      
Chain Protein ID
      
Chain Protein Compound
    
  
  

    

      
45
      
11
      
3.10
      
6JHE
      
2020-01-01
      
3.10
      
b
      
dna (5'-d(p*tp*tp*gp*ap*ap*ap*cp*cp*tp*tp*t)-3')
      
c
      
dna (5'-d(*ap*ap*ap*gp*gp*tp*tp*tp*cp*ap*a)-3')
      
a
      
ecf rna polymerase sigma factor sigw
    
    

      
59
      
0
      
1.90
      
6QEC
      
2020-02-05
      
1.90
      
u
      
dna (5'-d(*ap*tp*tp*cp*gp*ap*ap*tp*ap*t*tp*ap*...
      
b
      
dna (5'-d(*ap*tp*tp*cp*gp*ap*ap*tp*ap*t*tp*ap*...
      
a
      
transcription factor lux
    
    

      
7
      
0
      
2.70
      
4S2Q
      
2016-02-17
      
2.70
      
a
      
dna (5'-d(p*ap*gp*gp*cp*tp*tp*tp*gp*tp*tp*cp*t...
      
b
      
dna (5'-d(p*ap*gp*gp*ap*gp*ap*ap*cp*ap*ap*ap*g...
      
d
      
transcription factor sox-9
    
    

      
50
      
7
      
2.75
      
5JLX
      
2016-06-22
      
2.75
      
b, e
      
dna (5'-d(*ap*gp*ap*ap*ap*gp*cp*(c7s)p*ap*tp*t...
      
c, f
      
dna (5'-d(*tp*cp*tp*cp*tp*ap*ap*tp*gp*gp*cp*tp...
      
a, d
      
homeotic protein antennapedia
    
    

      
92
      
3
      
2.40
      
3U2B
      
2011-12-28
      
2.40
      
a
      
dna (5'-d(*gp*tp*cp*tp*cp*tp*ap*tp*tp*gp*tp*cp...
      
b
      
dna (5'-d(*cp*cp*ap*gp*gp*ap*cp*ap*ap*tp*ap*gp...
      
c
      
transcription factor sox-4
    
    

      
27
      
6
      
1.45
      
5KL3
      
2016-09-14
      
1.45
      
b
      
dna (5'-d(*ap*gp*cp*gp*tp*gp*gp*gp*ap*gp*t)-3')
      
c
      
dna (5'-d(*tp*ap*cp*tp*cp*cp*cp*ap*cp*gp*c)-3')
      
a
      
wilms tumor protein
    
  

In [ ]: