count lines with same value in column in python

问题

I'm trying to reproduce the R aggregate() function in python but without concatenating. For each line, I just want to count the number of occurrences of lines with a similar value in a given column.

I'm trying to work it out from a piece of code taken here: http://timotheepoisot.fr/2011/12/01/the-aggregate-function-in-python/

The modifications I implemented are indicated by ###. The problem I am currently having is that the first column [0] contains character strings and the code seems to work only with floats.

import numpy as np
import scipy as sp  

def MSD(vec):
    return [np.mean(vec),np.std(vec)]  
def aggregate(df,by=0,to=1,func=np.sum):
    Dat = []
#    ColBy = df.T[by]
    ColBy = int(df.T[by][3:]) ### my attempt to read only the numbers in the first column's character strings
    ColTo = df.T[to] 
    UniqueBy = np.sort(np.unique(ColBy))
    for ub in UniqueBy:
        uTo = ColTo[ColBy==ub]
        Out = func(uTo)
#        Dat.append(np.concatenate(([ub],Out)))
        Dat.append([ub],Out) ### because I do not want to concatenate
    return Dat  

test_df = np.loadtxt('in_test.txt')  
Agr = aggregate(test_df,0,3,MSD)
sp.savetxt("out_test.txt", Agr)

This is the error message:

Traceback (most recent call last):
  File "count_same_reads.py", line 30, in <module>
    test_df = np.loadtxt('in_test.txt')
  File "/usr/lib/python2.7/dist-packages/numpy/lib/npyio.py", line 796, in loadtxt
    items = [conv(val) for (conv, val) in zip(converters, vals)]
ValueError: could not convert string to float: Tag19184

My data is tab-delimited, containing mostly strings, except for column 3 in which I want to write the number of occurrences of lines.

Here is the test data:

Tag19184    CTAAC   hffef   1   a   36  -   chr1    10006   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10012   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10018   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10024   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10030   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10036   0   36M 36
Tag19184    CTAAC   hffef   1   a   36  -   chr1    10042   0   36M 36
Tag20198    CTAAC   hffef   1   a   36  -   chr1    10048   0   36M 36
Tag20198    CTAAC   hffef   1   a   36  -   chr1    10054   0   36M 36
Tag45093    CTAAC   hffef   1   a   36  -   chr1    10060   0   36M 36

The result should look like this:

Tag19184    CTAAC   hffef   7   a   36  -   chr1    10006   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10012   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10018   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10024   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10030   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10036   0   36M 36
Tag19184    CTAAC   hffef   7   a   36  -   chr1    10042   0   36M 36
Tag20198    CTAAC   hffef   2   a   36  -   chr1    10048   0   36M 36
Tag20198    CTAAC   hffef   2   a   36  -   chr1    10054   0   36M 36
Tag45093    CTAAC   hffef   1   a   36  -   chr1    10060   0   36M 36

As you can probably tell, I'm not so good at python yet. Any advice would be welcome.

[EDIT] PS. The data is already sorted by column [0].

回答1:

I will suggest pandas, especially in your case of genomic data, the size of the data may be quite large:

In [44]:
#you can read you data by pandas.read_csv()
import pandas as pd
print df
         v0     v1     v2  v3 v4  v5 v6    v7     v8  v9  v10  v11
0  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10006   0  36M   36
1  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10012   0  36M   36
2  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10018   0  36M   36
3  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10024   0  36M   36
4  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10030   0  36M   36
5  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10036   0  36M   36
6  Tag19184  CTAAC  hffef   1  a  36  -  chr1  10042   0  36M   36
7  Tag20198  CTAAC  hffef   1  a  36  -  chr1  10048   0  36M   36
8  Tag20198  CTAAC  hffef   1  a  36  -  chr1  10054   0  36M   36
9  Tag45093  CTAAC  hffef   1  a  36  -  chr1  10060   0  36M   36
In [45]:
#if we want to group by the first 3 fields
df.groupby(['v0','v1','v2']).transform(sum).v3
Out[45]:
0    7
1    7
2    7
3    7
4    7
5    7
6    7
7    2
8    2
9    1
Name: v3, dtype: int64
In [46]:
#all it takes is just one line
df['v3']=df.groupby(['v0','v1','v2']).transform(sum).v3
print df
         v0     v1     v2  v3 v4  v5 v6    v7     v8  v9  v10  v11
0  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10006   0  36M   36
1  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10012   0  36M   36
2  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10018   0  36M   36
3  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10024   0  36M   36
4  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10030   0  36M   36
5  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10036   0  36M   36
6  Tag19184  CTAAC  hffef   7  a  36  -  chr1  10042   0  36M   36
7  Tag20198  CTAAC  hffef   2  a  36  -  chr1  10048   0  36M   36
8  Tag20198  CTAAC  hffef   2  a  36  -  chr1  10054   0  36M   36
9  Tag45093  CTAAC  hffef   1  a  36  -  chr1  10060   0  36M   36

来源：https://stackoverflow.com/questions/24581967/count-lines-with-same-value-in-column-in-python

标签

python

count

concatenation

aggregate