Python Programming for Biology: Bioinformatics and Beyond

Figure 6.1.  An example of position (column) formatted data. This extract is from a

Protein Data Bank file which is used to represent data about molecules and their three-

dimensional structures.

---

Figure 6.2.  An example of a plain text format that uses keys and values. Here the

last value extends over multiple lines. This extract is a fragment of a file format called

mmCIF. As with the PDB format mmCIF is used to represent data about biological

molecules and their structure.

Figure 6.3.  An example of a file format containing tagged elements. A truncated

fragment of an XML file from the Protein Data Bank (PDBML).

Reading files

Using ‘

open

’

When reading a file within a Python program, you obviously need to specify the location

and  the  name  of  the  file,  collectively  termed  the  path  to  the  file.  The  file’s  path  is  then

used by the inbuilt open() function, or similar, to get access to the file’s contents. The path

can be absolute, which is to say that it starts at the top or root of the operating-system file

hierarchy, or the path can be relative to the current working directory (folder). Initially the

current working directory is the directory where the Python interpreter starts, although it

can be changed from inside the code if needed. In Python the file path is just a string, so,

for example, you could have:

path = 'examples/dataFile.txt'

fileObj = open(path)

The  open  function  passes  back  what  is  termed  a  file  handle,  a  Python  object  used  to

access  and  represent  the  open  file.  Here  the  file  handle  is  stored  in  a  variable  called

fileObj.  It  might  be  tempting  to  use  the  more  natural  variable  name  ‘file’,  but  in  Python

the word ‘file’ is already used to describe the type (class) of such objects. Hence, it is best

to  avoid  overriding  the  internal  variable  name.  An  alternative  to  the  descriptive  but

verbose ‘fileObj’ is the shorter ‘fh’, meaning ‘file handle’.

---

There is the possibility that a program may fail to open the file with the specified path.

For example, the stated file may simply not exist (e.g. the name was wrong), or you may

not  have  permission  to  read  it.  Under  such  circumstances,  when  you  try  to  open  it  the

function will throw an error, a standard Python exception (IOError). Otherwise, if all goes

well,  when  you  are  done  reading  a  file  then  you  can  explicitly  say  you  are  finished,  by

using  the  close()  function,  a  method  that  is  known  to  the  file  object  class.  A  listing  of

methods associated with file objects is given in

Appendix 2

.

fileObj.close()

Usually  you  do  not  have  to  explicitly  close  a  file,  because  when  the  variable  name

fileObj  goes  out  of  scope  and  is  no  longer  meaningful,  such  as  at  the  end  of  a  loop  or

function,  Python  will  automatically  close  it.

2

 However,  it  is  generally  a  good  habit  to

explicitly  close  a  file  handle  when  you  know  you  are  done,  just  in  case  it  is  important.

Also, this lets someone who reads your code know that you have finished using the file.

The  open()  function  can  take  an  optional  second  argument:  a  ‘mode’,  specifying  how

the file should be opened. These modes include ‘r’ to read from a file and ‘w’ to write to a

file, but, as you might have spotted in the above examples, the ‘r’ is optional, given that it

is the default. Although we could still have explicitly stated reading mode:

fileObj = open(path, 'r')