create_input.py

# Dictionary saving the def2-SVPD basis set.
# The code for generating the input file is down below this dictionary.
# https://stackoverflow.com/questions/61465411/save-a-multiple-lines-in-a-dict-in-python-question
def2SVPD_basis = {
    1 : r"""H     0
S   3   1.00
     13.0107010              0.19682158D-01
      1.9622572              0.13796524
      0.44453796             0.47831935
S   1   1.00
      0.12194962             1.0000000
P   1   1.00
      0.8000000              1.0000000
P   1   1.00
      0.11704099050          1.0000000
****""",
    6 : r"""C     0
S   5   1.00
   1238.4016938              0.54568832082D-02
    186.29004992             0.40638409211D-01
     42.251176346            0.18025593888
     11.676557932            0.46315121755
      3.5930506482           0.44087173314
S   1   1.00
      0.40245147363          1.0000000
S   1   1.00
      0.13090182668          1.0000000
S   1   1.00
      0.67053540256D-01      1.0000000
P   3   1.00
      9.4680970621           0.38387871728D-01
      2.0103545142           0.21117025112
      0.54771004707          0.51328172114
P   1   1.00
      0.15268613795          1.0000000
D   1   1.00
      0.8000000              1.0000000
D   1   1.00
      0.11713185140          1.0000000
****""",
    7 : r"""N     0
S   5   1.00
   1712.8415853             -0.53934125305D-02
    257.64812677            -0.40221581118D-01
     58.458245853           -0.17931144990
     16.198367905           -0.46376317823
      5.0052600809          -0.44171422662
S   1   1.00
      0.58731856571          1.0000000
S   1   1.00
      0.18764592253          1.0000000
S   1   1.00
      0.96171241529D-01      1.0000000
P   3   1.00
     13.571470233           -0.40072398852D-01
      2.9257372874          -0.21807045028
      0.79927750754         -0.51294466049
P   1   1.00
      0.21954348034          1.0000000
D   1   1.00
      1.0000000              1.0000000
D   1   1.00
      0.16697708112          1.0000000
****""",
    8 : r"""O     0
S   5   1.00
   2266.1767785             -0.53431809926D-02
    340.87010191            -0.39890039230D-01
     77.363135167           -0.17853911985
     21.479644940           -0.46427684959
      6.6589433124          -0.44309745172
S   1   1.00
      0.80975975668          1.0000000
S   1   1.00
      0.25530772234          1.0000000
S   1   1.00
      0.76572453250D-01      1.0000000
P   3   1.00
     17.721504317            0.43394573193D-01
      3.8635505440           0.23094120765
      1.0480920883           0.51375311064
P   1   1.00
      0.27641544411          1.0000000
P   1   1.00
      0.69002276350D-01      1.0000000
D   1   1.00
      1.2000000              1.0000000
D   1   1.00
      0.17992024323          1.0000000
****""",
    9 : r"""F     0
S   5   1.00
   2894.8325990             -0.53408255515D-02
    435.41939120            -0.39904258866D-01
     98.843328866           -0.17912768038
     27.485198001           -0.46758090825
      8.5405498171          -0.44653131020
S   1   1.00
      1.0654578038           1.0000000
S   1   1.00
      0.33247346748          1.0000000
S   1   1.00
      0.98097752264D-01      1.0000000
P   3   1.00
     22.696633924           -0.45212874436D-01
      4.9872339257          -0.23754317067
      1.3491613954          -0.51287353587
P   1   1.00
      0.34829881977          1.0000000
P   1   1.00
      0.83372148304D-01      1.0000000
D   1   1.00
      1.4000000              1.0000000
D   1   1.00
      0.22301361948          1.0000000
****""",
    15 : r"""P     0
S   5   1.00
   8002.4795106              0.57503489302D-02
   1203.6813590              0.43007628959D-01
    273.44227031             0.19363986252
     76.655541517            0.49651693399
     23.516927435            0.44983262479
S   3   1.00
     39.791683439            0.95188129789D-01
      4.2770343323          -0.57649840368
      1.6940256888          -0.54239583865
S   1   1.00
      0.27567674644          1.0000000
S   1   1.00
      0.10495590099          1.0000000
S   1   1.00
      0.64309734755D-01      1.0000000
P   5   1.00
    219.50755823             0.92100565257D-02
     51.274155030            0.65409765745D-01
     15.921595892            0.24033730279
      5.5069913481           0.46318321788
      1.9537719426           0.37392563382
P   1   1.00
      0.47803397932          1.0000000
P   1   1.00
      0.13657952621          1.0000000
D   1   1.00
      0.4500000              1.0000000
D   1   1.00
      0.99587311089D-01      1.0000000
****""",
    16 : r"""S     0
S   5   1.00
   9184.9303010             -0.22294387756D-02
   1381.5105503             -0.16683029937D-01
    313.87147580            -0.75262436116D-01
     88.053870623           -0.19376827038
     27.039914905           -0.17718020803
S   3   1.00
     45.648731303           -0.10736062573
      4.9664522326           0.65066293018
      2.0116242047           0.59712155354
S   1   1.00
      0.35661077013          1.0000000
S   1   1.00
      0.13507221477          1.0000000
S   1   1.00
      0.44999823506D-01      1.0000000
P   5   1.00
    261.98233439            -0.92729929822D-02
     61.306894736           -0.66547669241D-01
     19.103729887           -0.24828595903
      6.6567720378          -0.48703847402
      2.3959635161          -0.39337850312
P   1   1.00
      0.61776161679          1.0000000
P   1   1.00
      0.16993376871          1.0000000
P   1   1.00
      0.41706793676D-01      1.0000000
D   1   1.00
      0.5500000              1.0000000
D   1   1.00
      0.10172853717          1.0000000
****""",
    17 : r"""Cl     0
S   5   1.00
  10449.8275660              0.19708362484D-02
   1571.7365221              0.14754727977D-01
    357.12065523             0.66679112875D-01
    100.25185935             0.17228924084
     30.812727554            0.15883786100
S   3   1.00
     51.923789434           -0.10009298909
      5.7045760975           0.60841752753
      2.3508376809           0.54352153355
S   1   1.00
      0.44605124672          1.0000000
S   1   1.00
      0.16848856190          1.0000000
S   1   1.00
      0.55344229413D-01      1.0000000
P   5   1.00
    307.66790569            -0.87801484118D-02
     72.102015515           -0.63563355471D-01
     22.532680262           -0.24016428276
      7.8991765444          -0.47798866557
      2.8767268321          -0.38515850005
P   1   1.00
      0.77459363955          1.0000000
P   1   1.00
      0.21037699698          1.0000000
P   1   1.00
      0.50046271086D-01      1.0000000
D   1   1.00
      0.6500000              1.0000000
D   1   1.00
      0.12284803390          1.0000000
****""",
    34 : r"""Se     0
S   6   1.00
 106612.2002700              0.14274889113D-02
  16007.6047010              0.10943525114D-01
   3642.1699707              0.54374171596D-01
   1028.5912993              0.19018092947
    335.30298888             0.38913021696
    115.80129154             0.30620207088
S   3   1.00
    222.93325020            -0.11198808009
     26.726257934            0.64752124207
     11.124501923            0.44241976651
S   3   1.00
     19.888520061           -0.22857227762
      3.3668473803           0.73591359951
      1.5249277839           0.44330199577
S   1   1.00
      0.29630037912          1.0000000
S   1   1.00
      0.11009288974          1.0000000
S   1   1.00
      0.42167294728D-01      1.0000000
P   5   1.00
   1455.9068120              0.88203597043D-02
    343.75101831             0.66875851967D-01
    109.29554964             0.26640578080
     39.707711022            0.53834928422
     14.950185232            0.36303281993
P   3   1.00
      8.0208962094           0.34153807025
      3.2934649756           0.57257906583
      1.4058602438           0.25549813222
P   1   1.00
      0.45076123226          1.0000000
P   1   1.00
      0.13353413325          1.0000000
P   1   1.00
      0.34396697938D-01      1.0000000
D   4   1.00
     94.494024044            0.23490101098D-01
     27.188185260            0.13747735767D+00
      9.5091567352           0.36649929124D+00
      3.4170516853           0.48750989884D+00
D   1   1.00
      1.1479590083            .27657943383D+00
D   1   1.00
      0.3380000              1.0000000
D   1   1.00
      0.82767380906D-01      1.0000000
****""",
    35 : r"""Br     0
S   6   1.00
 113286.3877600              0.14283037779D-02
  17009.6263030              0.10950417496D-01
   3870.1842567              0.54421006604D-01
   1093.0357227              0.19047907695
    356.39721797             0.39024642737
    123.12539643             0.30814432514
S   3   1.00
    236.74084007            -0.11228065671
     28.468661070            0.64775962312
     11.883443722            0.44235575986
S   3   1.00
     21.269633312           -0.22642576323
      3.6129226841           0.73823712008
      1.6626648969           0.42683868694
S   1   1.00
      0.34823793232          1.0000000
S   1   1.00
      0.13019031394          1.0000000
S   1   1.00
      0.50966709763D-01      1.0000000
P   5   1.00
   1560.2801881              0.87166669072D-02
    368.47859205             0.66243637420D-01
    117.22978849             0.26495610385
     42.648909248            0.53839160587
     16.087225096            0.36579387888
P   3   1.00
      8.6352810058           0.34248787366
      3.5613665502           0.57500678213
      1.5292626609           0.24330394172
P   1   1.00
      0.53064294848          1.0000000
P   1   1.00
      0.15702758965          1.0000000
P   1   1.00
      0.39636841846D-01      1.0000000
D   4   1.00
    104.85518642             0.22650147581D-01
     30.281143688            0.13455483230
     10.651394267            0.36474454537
      3.8699456233           0.49044587056
D   1   1.00
      1.3240876762            .27137289040
D   1   1.00
      0.3890000              1.0000000
D   1   1.00
      0.96543342393D-01      1.0000000
****""",
    53 : r"""I     0
S   6   1.00
    445.90489176             0.20037290389D-02
     23.336842412           -0.14989397324
     19.583446104            0.36436474913
      8.5112089186          -0.75538138199
      2.1996161861           0.82816359356
      1.0668970454           0.42161048110
S   2   1.00
     11.720547572           -0.16160234511D-01
      1.7625986449           0.35271525193
S   1   1.00
      0.28697476318          1.0000000
S   1   1.00
      0.11227609607          1.0000000
S   1   1.00
      0.45123635492D-01      1.0000000
P   4   1.00
     20.499027254            0.77159335656D-01
     10.558754576           -0.38370372780
      1.5015895485           0.83554174719
      0.64597173095          0.16787259488
P   1   1.00
      3.0288656771           1.0000000
P   1   1.00
      0.34506591841          1.0000000
P   1   1.00
      0.11134513813          1.0000000
P   1   1.00
      0.29795507230D-01      1.0000000
D   5   1.00
     51.235354920            0.40345130346D-02
     15.616239482           -0.50768108881D-02
      4.5266002139           0.29151065356
      2.0529808766           0.51145783605
      0.87640281623          0.31232025297
D   1   1.00
      0.30900000000          1.0000000
D   1   1.00
      0.77987255180D-01      1.0000000
****

I     0
I-ECP     3     28
f potential
  4
2     19.45860900           -21.84204000
2     19.34926000           -28.46819100
2      4.82376700            -0.24371300
2      4.88431500            -0.32080400
s-f potential
  7
2     40.01583500            49.99429300
2     17.42974700           281.02531700
2      9.00548400            61.57332600
2     19.45860900            21.84204000
2     19.34926000            28.46819100
2      4.82376700             0.24371300
2      4.88431500             0.32080400
p-f potential
  8
2     15.35546600            67.44284100
2     14.97183300           134.88113700
2      8.96016400            14.67505100
2      8.25909600            29.37566600
2     19.45860900            21.84204000
2     19.34926000            28.46819100
2      4.82376700             0.24371300
2      4.88431500             0.32080400
d-f potential
  10
2     15.06890800            35.43952900
2     14.55532200            53.17605700
2      6.71864700             9.06719500
2      6.45639300            13.20693700
2      1.19177900             0.08933500
2      1.29115700             0.05238000
2     19.45860900            21.84204000
2     19.34926000            28.46819100
2      4.82376700             0.24371300
2      4.88431500             0.32080400
****"""
}

# ------------------------------------------
# CODE FOR AUTOMATING INPUT FILE GENERATION
# ------------------------------------------

# This code only works for the wB97X-D/def2-SVPD model chemistry.

# To run this script the following files are needed:
    # - molecules.txt: contains a list with all molecules that will be considering
    # - vpt2_def2.gjf: input file template for VPT2 jobs with wB97X-D/def2-SVPD model chemistry in Gaussian
    # - subtem.pbs: template submission file
    # - geomtries folder: constains all geometries (in z-matrix) for the molecules listed in molecules.txt

import os
import re
import shutil
from mendeleev import element

failed_jobs = []

# Creates outputs folder
# https://stackabuse.com/creating-and-deleting-directories-with-python/
os.mkdir('outputs')

# Reads molecules.txt file and loops over all molecules present in the file.
with open('molecules.txt','r') as inp_molecules:
    for molecule in inp_molecules:
        
        try:
            
            basis_data = []
            elements_list = []
            atomic_numbers = []
            basis_str = ''
        
            # Moves to outputs folder and copies the template file, changing its name to the one
            # for the molecule being considered.
            # https://thispointer.com/python-how-to-copy-files-from-one-location-to-another-using-shutil-copy/
            os.chdir('outputs')
            shutil.copy('../vpt2_def2.gjf', molecule.rstrip()+'_vpt2.gjf')
        
            # Reads the elements present in the molecular formula and saves them into the elemens_list.
            for letter in re.findall('[A-Z][^A-Z]*', molecule.rstrip()):
                symbol = re.split(r'[0-9]', letter.split()[0])[0]
                poss_under = re.split(r'(_)', symbol.split()[0])[0]
                elements_list.append(poss_under)
        
            # Creates a new list where stores the atomic numbers for each element present in the molecular formula.
            # The list is then sorted from the smallers to the largest value.
            for atom in elements_list:
                symbol = element(atom)
                atomic_numbers.append(symbol.atomic_number)
            
            atomic_numbers.sort()
        
            # Reads over each atomic number and saves the corresponding basis set for each element in the basis_data list.
            # The basis_data is organised with increasing atomic number, as required by Gaussian for the input file.
            for number in atomic_numbers:
                if number in def2SVPD_basis:
                    basis_data.append(def2SVPD_basis[number])
                
            # The basis is joined into a single string to be pasted in the input file template.
            basis_str = '\n'.join(basis_data)
        
            # Saves the info for the respective geometry into the geom_data list.
            with open('../10_nISeBr/'+molecule.rstrip()+'.zmax','r') as inp_geom:
                geom_data = inp_geom.read()
                inp_geom.close()
        
            # Replaces the selected keywords (i.e. {molecule}, {geometry} and {def2_basis}) with the
            # corresponding values.
            # https://www.kite.com/python/answers/how-to-update-and-replace-text-in-a-file-in-python
            with open(molecule.rstrip()+'_vpt2.gjf','r+') as inp_template:
                template_data = inp_template.read()
                template_data = re.sub('{geometry}', geom_data, template_data)
                template_data = re.sub('{molecule}', molecule.rstrip(), template_data)
                template_data = re.sub('{def2_basis}', basis_str, template_data)

                # writes the new templates.
                inp_template.seek(0)
                inp_template.write(template_data)
                inp_template.truncate()
        
            # Copies submission file template and pastes the appropriate input file name.
            shutil.copy('../subtem.pbs', molecule.rstrip()+'_vpt2.pbs')
            with open(molecule.rstrip()+'_vpt2.pbs','r+') as inp_sub:
                submission_data = inp_sub.read()
                submission_data = re.sub('{file_name}', molecule.rstrip()+'_vpt2', submission_data)
            
                inp_sub.seek(0)
                inp_sub.write(submission_data)
                inp_sub.truncate()
                
        except:
            failed_jobs.append(molecule.rstrip())
            
        os.chdir('../')