In order to convert files in ConKit, we need to import the ConKit I/O framework.
>>> from conkit import io
Note
ConKit I/O framework consists of three main functions that handle the relevant parsers: conkit.io.read(), conkit.io.write() and conkit.io.convert(). The latter effectively uses the former but handles everything in one step.
Let’s say you are designing a new pipeline and you want to manipulate your contact file. Thus, you need to have a way of reading the file into memory before you can manipulate it.
ConKit provides the following simple syntax to do exactly that.
>>> with open('query.pconsc3.txt', 'r') as f_in:
... contact_hierarchy = io.read(f_in, 'pconsc3')
Now, you’ve manipulated the file from the previous example and you now want to write it back out. Look no further, this is what you need to do.
>>> with open('query.pconsc3.new.txt', 'w') as f_out:
... io.write(f_out, contact_hierarchy, 'pconsc3')
Imagine the following scenario: you have just used an online server, let’s say PconsC3, and you have received a contact prediction file. Now, the file in text format is little use to you and although you have a contact map, there are many cool tools that use predictions to visualise them in a protein structure. These tools rely on the official Casp format and you really don’t want to convert every line manually or write a script to do it. Look no further, ConKit does this in a heartbeat.
Assuming our PconsC3 contact file is called query.pconsc3.txt and you want to write it to a file called query.rr.txt, you can convert it using the following code:
>>> with open('query.pconsc3.txt', 'r') as f_in, open('query.casp.rr', 'w') as f_out:
... io.convert(f_in, 'pconsc3', f_out, 'casprr')
This is it, you now have converted your contact file from PconsC3 format to Casp format.
The functions used to convert sequence files are identical to those used for contact prediction files. The only difference is the format parsed to those functions.