如何将两个一维列表组合成二维数组？

Question

I am trying to read mzXML files using Pyteomics' mzxml class. The elements that I need to access are in numpy.ndarray format, which I convert as lists. The mzXML files contain several columns with lists as values. The main objective is to combine the two lists into 2D array (side by side in column-wise) so that I can save as CSV files.

I tried using np.concatenate((mzplist, mzplist2), axis=1) , which produced axis=1 error saying that axis=1 is out of bounds for 1D arrays. I also tried using hstack , column_stack . The closest I got was from column_stack (code below) but the resulting array was 1D when I viewed the resulting CSV files (each cell of Excel contains m/z value and intensity value separated by a space).

plist = []

for files in os.listdir(full_path):
    filename = os.path.basename(files)
    with mzxml.read(full_path + '\\' + filename) as reader:
        for line in reader:
            mzplist = line['m/z array'].tolist()
            mzplist2 = line['intensity array'].tolist()
            print(type(mzplist))
            mzplist = np.column_stack([mzplist, mzplist2])
            #mzplist.columns = ['mass', 'Intensity']
            np.savetxt(newfolder + '\\' + filename + '.csv', mzplist) 
            plist = []
            mzplist = []
            mzplist2 = []

Expected results for mzplist :

 Mass       Intensity
  1            2
  3            4
  5            6

Here line['m/z array'].tolist() yields a list [1, 3, 5, ...] , and line['intensity array'].tolist() yields a list [2, 4, 6, ...] .

Am I missing something?

Answer 1

each cell of Excel contains m/z value and intensity value separated by a space

I suspect problem source is that line

np.savetxt(newfolder + '\\' + filename + '.csv', mzplist)

as space is default delimiter for np.savetxt ( as documentation say ), try to replace that line with

np.savetxt(newfolder + '\\' + filename + '.csv', mzplist, delimiter=',')

and check if that would help.

Answer 2

With 2 lists as you describe:

In [39]: alist=[1,3,5,7]; blist=[2,4,6,8]

A natural way to combine them into an array is:

In [40]: arr = np.array((alist, blist))
In [41]: arr
Out[41]: 
array([[1, 3, 5, 7],
       [2, 4, 6, 8]])

Transpose of that array looks like:

In [42]: arr.T
Out[42]: 
array([[1, 2],
       [3, 4],
       [5, 6],
       [7, 8]])

Which we can write with savetxt as:

In [44]: np.savetxt('foo.txt', arr.T, fmt='%5d')
In [45]: cat foo.txt
    1     2
    3     4
    5     6
    7     8

column_stack and c_ will produce the same array.

You can add a ',' delimiter if that is what your external reader demands.

Do you know how to read the output of a savetxt write as plain text? I'm using the bash shell cat .

When people have problems reading and writing csv files we usually ask for samples, so we can reproduce the problem. If needed a sample of intermediate arrays (such as the output of the column_stack ) may help. Otherwise we are left guessing as to what the problem is.