Python function about chemical formulas-CodePudding

I have a CSV file that contains chemical matter names and some info.What I need to do is add new columns and write their formulas, molecular weights and count H,C,N,O,S atom numbers in each formula.I am stuck with the counting atom numbers part.I have the function related it but I don't know how to merge it and make code work.

import pandas as pd    
import urllib.request    
import copy    
import re    

df = pd.read_csv('AminoAcids.csv')

def countAtoms(string, dict={}):
    curDict = copy.copy(dict)
    atoms = re.findall("[A-Z]{1}[a-z]*[0-9]*", string)

    for j in atoms:
        atomGroups = re.match('([A-Z]{1}[a-z]*)([0-9]*)', j)
        atom = atomGroups.group(1)
        number = atomGroups.group(2)
        try :
            curDict[atom] = curDict[atom]   int(number)
        except KeyError:
            try :
                curDict[atom] = int(number)
            except ValueError:
                curDict[atom] = 1
        except ValueError:
            curDict[atom] = curDict[atom]   1
    return curDict

df["Formula"] = ['C3H7NO2', 'C6H14N4O2 ','C4H8N2O3','C4H7NO4 ',
'C3H7NO2S ','C5H9NO4','C5H10N2O3','C2H5NO2 ','C6H9N3O2',
'C6H13NO2','C6H13NO2','C6H14N2O2 ','C5H11NO2S ','C9H11NO2',
'C5H9NO2 ','C3H7NO3','C4H9NO3 ','C11H12N2O2 ','C9H11NO3 ','C5H11NO2']
df["Molecular Weight"] = ['89.09','174.2','132.12',
'133.1','121.16','147.13','146.14','75.07','155.15',
'131.17','131.17','146.19','149.21','165.19','115.13',
'105.09','119.12','204.22','181.19','117.15']
df["H"] = 0
df["C"] = 0
df["N"] = 0
df["O"] = 0
df["S"] = 0
df.to_csv("AminoAcids.csv", index=False)
print(df.to_string())

CodePudding user response：

If I understand correctly, you should be able to use str.extract here:

df["H"] = df["Formula"].str.extract(r'H(\d )')
df["C"] = df["Formula"].str.extract(r'C(\d )')
df["N"] = df["Formula"].str.extract(r'N(\d )')
df["O"] = df["Formula"].str.extract(r'O(\d )')
df["S"] = df["Formula"].str.extract(r'S(\d )')

CodePudding user response：

here is another approach with similar result:

df.join(df['Formula'].str.findall('([A-Z])(\d*)').map(dict).apply(pd.Series).replace('', 1))

>>>
'''
        Formula Molecular Weight   C   H  N  O    S
0       C3H7NO2            89.09   3   7  1  2  NaN
1    C6H14N4O2             174.2   6  14  4  2  NaN
2      C4H8N2O3           132.12   4   8  2  3  NaN
3      C4H7NO4             133.1   4   7  1  4  NaN
4     C3H7NO2S            121.16   3   7  1  2  1.0
5       C5H9NO4           147.13   5   9  1  4  NaN
6     C5H10N2O3           146.14   5  10  2  3  NaN
7      C2H5NO2             75.07   2   5  1  2  NaN
8      C6H9N3O2           155.15   6   9  3  2  NaN
9      C6H13NO2           131.17   6  13  1  2  NaN
10     C6H13NO2           131.17   6  13  1  2  NaN
11   C6H14N2O2            146.19   6  14  2  2  NaN
12   C5H11NO2S            149.21   5  11  1  2  1.0
13     C9H11NO2           165.19   9  11  1  2  NaN
14     C5H9NO2            115.13   5   9  1  2  NaN
15      C3H7NO3           105.09   3   7  1  3  NaN
16     C4H9NO3            119.12   4   9  1  3  NaN
17  C11H12N2O2            204.22  11  12  2  2  NaN
18    C9H11NO3            181.19   9  11  1  3  NaN
19     C5H11NO2           117.15   5  11  1  2  NaN