notebook.community

Edit and run 





In [1]:



    


import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import datetime

sns.set(rc={'figure.figsize':(10,6.180)})
sns.set_style("whitegrid")

%matplotlib inline



    











In [106]:



    


def my_reorder(a, first):
    # move first to the top. and keep the rest
    new_order = first.copy()
    for col in a:
        if col not in first:
            new_order.append(col)
    return new_order

def read_pdb(pre, name, run=30, rerun=2):
    all_data = []
    for i in range(run):
        for j in range(rerun):
            # pre = "/Users/weilu/Research/server/nov_2018/iterative_optimization_4/all_simulations/"
            location = pre + f"{name}/simulation/{i}/{j}/"
            try:
                wham = pd.read_csv(location+"wham.dat")
            except:
                print(f"PDB: {name}, Run: {i}, Rerun: {j} not exist")
                print(location+"wham.dat")
                continue
            wham.columns = wham.columns.str.strip()
            remove_columns = ['Tc', 'Energy']
            wham = wham.drop(remove_columns, axis=1)
            energy = pd.read_csv(location+"energy.dat")
            energy.columns = energy.columns.str.strip()
            remove_columns = ['Steps', 'Shake', 'Excluded', 'Helix', 'AMH-Go', 'Vec_FM', 'SSB']
            energy = energy.drop(remove_columns, axis=1)
            data = pd.concat([wham, energy], axis=1).assign(Repeat=i, Run=j)
            all_data.append(data)
    data = pd.concat(all_data).reset_index(drop=True)
    data = data.reindex(columns=my_reorder(data.columns, ["Steps", "Qw", "VTotal", "Run", "Repeat"]))
    print(name, len(data))
    return data



    











In [49]:



    


dataset = {"old":"1R69, 1UTG, 3ICB, 256BA, 4CPV, 1CCR, 2MHR, 1MBA, 2FHA".split(", "),
            "new":"1FC2C, 1ENH, 2GB1, 2CRO, 1CTF, 4ICB".split(", "),
            "test":["t089", "t120", "t251", "top7", "1ubq", "t0766", "t0778", "t0782", "t0792", "t0803", "t0815", "t0833", "t0842", "t0844"]}
dataset["combined"] = dataset["old"] + dataset["new"]

def get_complete_data(pre, folder_list, pdb_list, **kwargs):
    complete_all_data = []
    for folder in folder_list:
        # pre = "/Users/weilu/Research/server/april_2019/iterative_optimization_old_set/"
        pre_folder = f"{pre}{folder}/"
        all_data = []
        for p in pdb_list:
            name = p.lower()[:4]
            tmp = read_pdb(pre_folder, name, **kwargs)
            all_data.append(tmp.assign(Name=name))
        data = pd.concat(all_data)
        complete_all_data.append(data.assign(Folder=folder))
    data = pd.concat(complete_all_data)
    data = data.reindex(columns=my_reorder(data.columns, ["Name", "Folder"]))
    return data



    











In [107]:



    


pre = "/Users/weilu/Research/server/april_2019/globular_2xov_named_2lep/"
folder_list = ["strengthen_beta"]
pdb_list = ["2lep"]
data = get_complete_data(pre, folder_list, pdb_list, run=20, rerun=1)
data_strength_beta = data



    


















    






2lep 40000

















In [108]:



    


pre = "/Users/weilu/Research/server/april_2019/globular_2xov_named_2lep/"
folder_list = ["longerRun"]
pdb_list = ["2lep"]
data = get_complete_data(pre, folder_list, pdb_list, run=20, rerun=2)
data_longerRun = data



    


















    






2lep 80000

















In [97]:



    


pre = "/Users/weilu/Research/server/april_2019/globular_2xov_named_2lep/"
folder_list = ["longerRun"]
pdb_list = ["2lep"]
data = get_complete_data(pre, folder_list, pdb_list, run=20, rerun=2)
# subset_data = data.query("Steps % 80000 == 0")
today = datetime.datetime.today().strftime('%m-%d')
print(today)
# data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")
# a = pd.read_csv("/Users/weilu/Research/data/optimization/energy_03-31.csv", index_col=0)
# b = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-01.csv", index_col=0)
# data = pd.concat([a,b])
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")



    


















    






2lep 80000
04-11

















In [101]:



    


data_strength_beta = data



    











In [99]:



    


data_longerRun = data



    











In [109]:



    


data = pd.concat([data_strength_beta, data_longerRun])

native water energy is about -53.





In [110]:



    


data.head(1)



    


















    
Out[110]:











  
    

      
      Name
      Folder
      Steps
      Qw
      VTotal
      Run
      Repeat
      Rg
      Chain
      Chi
      Rama
      DSSP
      P_AP
      Water
      Burial
      Frag_Mem
      Membrane
      Ebond
      Epair
    

  
  
    

      0
      2lep
      strengthen_beta
      4000
      0.095645
      -284.816011
      0
      0
      30.959568
      97.453826
      19.810282
      -103.119297
      0.0
      0.0
      0.05606
      -53.375831
      -245.64105
      0
      160.906666
      2.88755
    

  




















In [ ]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "DSSP", alpha=0.1).add_legend())



    











In [111]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "DSSP", alpha=0.1).add_legend())



    


















    
























In [104]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.1).add_legend())



    


















    
























In [98]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [62]:



    


pre = "/Users/weilu/Research/server/april_2019/iterative_optimization_combined_train_set/"
folder_list = ["iter2_normalized_noFrag"]
pdb_list = dataset["combined"]
data = get_complete_data(pre, folder_list, pdb_list, run=10, rerun=2)
subset_data = data.query("Steps % 80000 == 0")
today = datetime.datetime.today().strftime('%m-%d')
print(today)
# data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")
# a = pd.read_csv("/Users/weilu/Research/data/optimization/energy_03-31.csv", index_col=0)
# b = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-01.csv", index_col=0)
# data = pd.concat([a,b])
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")



    


















    






1r69 20000
1utg 20000
3icb 20000
256b 20000
4cpv 20000
1ccr 20000
2mhr 20000
1mba 15000
2fha 14421
1fc2 40000
1enh 40000
2gb1 40000
2cro 40000
1ctf 40000
4icb 40000
04-01

















In [72]:



    


data = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-01.csv", index_col=0)



    


















    






/Users/weilu/anaconda3/envs/py36/lib/python3.6/site-packages/numpy/lib/arraysetops.py:522: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison
  mask |= (ar1 == a)

















In [76]:



    


data.shape



    


















    
Out[76]:








(1228195, 17)

















In [84]:



    


pre = "/Users/weilu/Research/server/april_2019/iterative_optimization_combined_train_set_with_frag/"
folder_list = ["iter6_normalized_noFrag"]
pdb_list = dataset["combined"]
data = get_complete_data(pre, folder_list, pdb_list, run=10, rerun=2)
subset_data = data.query("Steps % 80000 == 0")
today = datetime.datetime.today().strftime('%m-%d')
print(today)
# data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")
# a = pd.read_csv("/Users/weilu/Research/data/optimization/energy_03-31.csv", index_col=0)
# b = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-01.csv", index_col=0)
# data = pd.concat([a,b])
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}_1.csv")



    


















    






1r69 20000
1utg 20000
3icb 20000
256b 20000
4cpv 20000
1ccr 20000
2mhr 20000
1mba 15000
2fha 15000
1fc2 40000
1enh 40000
2gb1 40000
2cro 40000
1ctf 40000
4icb 40000
04-08

















In [83]:



    


pre = "/Users/weilu/Research/server/april_2019/iterative_optimization_combined_train_set_with_frag/"
folder_list = ["iter0"]
pdb_list = dataset["combined"]
data = get_complete_data(pre, folder_list, pdb_list, run=30, rerun=1)
# subset_data = data.query("Steps % 80000 == 0")
today = datetime.datetime.today().strftime('%m-%d')
print(today)
# data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")
# a = pd.read_csv("/Users/weilu/Research/data/optimization/energy_03-31.csv", index_col=0)
# b = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-01.csv", index_col=0)
# data = pd.concat([a,b])
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")



    


















    






1r69 30000
1utg 30000
3icb 30000
256b 30000
4cpv 30000
1ccr 30000
2mhr 30000
1mba 22500
2fha 22500
1fc2 60000
1enh 60000
2gb1 60000
2cro 60000
1ctf 60000
4icb 60000
04-08

















In [112]:



    


data = pd.read_csv("/Users/weilu/Research/data/optimization/energy_04-08.csv", index_col=0)



    











In [113]:



    


subset_iter6 = subset_data



    











In [115]:



    


data.head()



    


















    
Out[115]:











  
    

      
      Name
      Folder
      Steps
      Qw
      VTotal
      Run
      Repeat
      Rg
      Chain
      Chi
      Rama
      P_AP
      Water
      Burial
      Frag_Mem
      Ebond
      Epair
    

  
  
    

      0
      1r69
      iter0
      4000
      0.342615
      -114.472189
      0
      0
      11.061036
      157.535308
      21.793236
      -48.797913
      0
      -10.796545
      -2.331958
      -231.874317
      284.228717
      11.066021
    

    

      1
      1r69
      iter0
      8000
      0.278288
      -203.720363
      0
      0
      12.400950
      117.297890
      38.255093
      -91.625458
      0
      -10.943550
      -4.078143
      -252.626195
      214.008935
      8.430595
    

    

      2
      1r69
      iter0
      12000
      0.279709
      -286.708326
      0
      0
      11.648768
      68.085923
      22.697867
      -105.052346
      0
      -6.586444
      -2.945326
      -262.907999
      181.378195
      3.043456
    

    

      3
      1r69
      iter0
      16000
      0.283629
      -331.560857
      0
      0
      11.789625
      99.347878
      19.623003
      -157.677660
      0
      -13.977479
      -3.824403
      -275.052196
      194.861902
      8.094861
    

    

      4
      1r69
      iter0
      20000
      0.302869
      -341.545376
      0
      0
      11.664210
      86.151590
      18.731492
      -153.732794
      0
      -7.723566
      -4.621929
      -280.350169
      162.867865
      3.861190
    

  




















In [116]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Frag_Mem", "Water", alpha=0.5).add_legend())



    


















    
























In [87]:



    


subset_iter0_with_rg = data.query("Steps % 80000 == 0")



    











In [92]:



    


data = pd.concat([subset_iter0_with_rg, subset_iter6])



    











In [ ]:

with rg constraint, we are able to find those unative collapsed structure.

I will add iter6 gamma, with constraint below.





In [95]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [93]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Rg", alpha=0.5).add_legend())



    


















    
























In [89]:



    


g = sns.FacetGrid(subset_iter0_with_rg, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Rg", alpha=0.5).add_legend())



    


















    
























In [88]:



    


g = sns.FacetGrid(subset_iter6, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Rg", alpha=0.5).add_legend())



    


















    
























In [82]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Rg", alpha=0.5).add_legend())



    


















    
























In [78]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Rg", alpha=0.5).add_legend())



    


















    
























In [75]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [64]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [50]:



    


pre = "/Users/weilu/Research/server/april_2019/iterative_optimization_combined_train_set/"
folder_list = ["iter0_normalized_noFrag", "iter1_normalized_noFrag"]
pdb_list = dataset["combined"]
data = get_complete_data(pre, folder_list, pdb_list, run=10, rerun=2)
subset_data = data.query("Steps % 80000 == 0")
today = datetime.datetime.today().strftime('%m-%d')
print(today)
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")



    


















    






1r69 20000
1utg 20000
3icb 20000
256b 20000
4cpv 20000
1ccr 20000
2mhr 20000
1mba 15000
2fha 14738
1fc2 40000
1enh 40000
2gb1 40000
2cro 40000
1ctf 40000
4icb 40000
1r69 20000
1utg 20000
3icb 20000
256b 20000
4cpv 20000
1ccr 20000
2mhr 20000
1mba 15000
2fha 14036
1fc2 40000
1enh 40000
2gb1 40000
2cro 40000
1ctf 40000
4icb 40000

















In [52]:



    


today = datetime.datetime.today().strftime('%m-%d')
print(today)
data.reset_index(drop=True).to_csv(f"/Users/weilu/Research/data/optimization/energy_{today}.csv")



    


















    






03-31

















In [57]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Steps", "Water", alpha=0.5).add_legend())



    


















    
























In [58]:



    


g = sns.FacetGrid(subset_data, col="Name",col_wrap=4,  hue="Folder", sharey=False, sharex=False)
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [ ]:



    






    











In [31]:



    


dataset = {"old":("1R69, 1UTG, 3ICB, 256BA, 4CPV, 1CCR, 2MHR, 1MBA, 2FHA".split(", "), 40),
            "new":("1FC2C, 1ENH, 2GB1, 2CRO, 1CTF, 4ICB".split(", "), 80),
            "test":(["t089", "t120", "t251", "top7", "1ubq", "t0766", "t0778", "t0782", "t0792", "t0803", "t0815", "t0833", "t0842", "t0844"], 40)}
pdb_list, steps = dataset["old"]

complete_all_data = []
for folder in ["single", "noFrag"]:
    pre = f"/Users/weilu/Research/server/april_2019/iterative_optimization_old_set/{folder}/"
    all_data = []
    for p in pdb_list:
        name = p.lower()[:4]
        tmp = read_pdb(pre, name, run=30, rerun=1)
        all_data.append(tmp.assign(Name=name))
    data = pd.concat(all_data)
    complete_all_data.append(data.assign(Folder=folder))
data = pd.concat(complete_all_data)
data = data.reindex(columns=my_reorder(data.columns, ["Name", "Folder"]))
data_subset = data.query("Steps % 80000 == 0")



    


















    






1r69 30000
1utg 30000
3icb 30000
256b 30000
4cpv 30000
1ccr 30000
2mhr 30000
1mba 23146
2fha 19885
1r69 30000
1utg 30000
3icb 30000
256b 30000
4cpv 30000
1ccr 30000
2mhr 30000
1mba 25680
2fha 22113

















In [32]:



    


data.head()



    


















    
Out[32]:











  
    

      
      Name
      Folder
      Steps
      Qw
      VTotal
      Run
      Repeat
      Rg
      Chain
      Chi
      Rama
      P_AP
      Water
      Burial
      Frag_Mem
      Ebond
      Epair
    

  
  
    

      0
      1r69
      single
      4000
      0.087288
      -97.248076
      0
      0
      30.620022
      84.805297
      28.447449
      -76.387581
      0
      -1.560166
      -51.343641
      -81.209434
      185.385749
      4.497708
    

    

      1
      1r69
      single
      8000
      0.179975
      -183.634765
      0
      0
      22.565086
      111.754883
      16.184323
      -113.969007
      0
      -4.159232
      -53.089330
      -140.356402
      177.924318
      14.831873
    

    

      2
      1r69
      single
      12000
      0.235191
      -220.810608
      0
      0
      16.929506
      95.902273
      21.480374
      -113.593651
      0
      -7.269499
      -53.867215
      -163.462890
      153.082652
      8.920596
    

    

      3
      1r69
      single
      16000
      0.377747
      -255.999037
      0
      0
      11.232515
      95.598735
      16.251947
      -133.302030
      0
      -22.810907
      -54.899661
      -156.837122
      178.698420
      4.564993
    

    

      4
      1r69
      single
      20000
      0.205453
      -211.454610
      0
      0
      18.008032
      91.545729
      19.502155
      -116.180706
      0
      -10.169564
      -52.802971
      -143.349253
      164.482353
      10.297213
    

  




















In [33]:



    


data_subset = data.query("Steps % 80000 == 0")



    











In [40]:



    


g = sns.FacetGrid(data_subset, col="Name",col_wrap=4,  hue="Folder")
g = (g.map(plt.scatter, "Qw", "Water", alpha=0.5).add_legend())



    


















    
























In [39]:



    


g = sns.FacetGrid(data_subset, col="Name",col_wrap=4,  hue="Folder")
g = (g.map(plt.scatter, "Steps", "Water", alpha=0.5).add_legend())



    


















    
























In [27]:



    


g = sns.FacetGrid(data_subset, col="Name",col_wrap=4)
g = g.map(plt.scatter, "Steps", "Frag_Mem")



    


















    
























In [29]:



    


g = sns.FacetGrid(data_subset, col="Name",col_wrap=4)
g = g.map(plt.scatter, "Steps", "Qw")



    


















    
























In [16]:



    


g = sns.FacetGrid(data, col="Name",col_wrap=4)
g = g.map(plt.scatter, "Steps", "Qw")



    


















    
























In [ ]:

Content source: luwei0917/awsemmd_script

Similar notebooks:

notebook.community | gallery | about