Take a string in the format:

data = ‘’’

1 4

2 8.2

3 12

4 15,16

’’’

and return the following list of numpy arrays: [np,array([1, 2, 3, 4]), np.array(4, 8.2, 12, 15.16)]

def from_string(data):
    import re
    data = re.sub(',','.', data)
    data = re.sub(r'\t',' ', data)
    data = re.sub(' +',' ', data)
    columns = []
    for line in data.splitlines():
        try:
            columns.append([float(x) for x in line.split(' ')])         
        except ValueError as e:
            pass
    return [np.array(raw) for raw in zip(*columns)]

Example:

data = '''

1  1
2  4.02
3  9
4  16.3
5 25

'''

x, y = from_string(data)
print(x - np.sqrt(y))

[ 0. -0.00499377 0. -0.03732585 0. ]