A test GitHub repository for a set of client/server python scripts (as part of the Scientific Programming class).
This script allows the user to send a request conversion:
- from DNA sequence to a Burrows-Wheeler Transform string
- from a Burrows-Wheeler Transform string to a DNA sequence The host:port required by flask (an external package https://flask.palletsprojects.com/en/2.3.x/) is manually inputed by the user when invoking the client-to-server connection. Here's an example of the syntax required:
python3 BWT_client.py method host port sequence
The method can either be bwt (i.e. DNA to BWT), or ibwt (i.e. inverse bwt, BWT to DNA), the host should be in the 0.0.0.0 format and port should be in the 10^3 order of magnitude.
Sequence can be inputed either directly as a string between "", or from an external .txt file: the script handles both approaches, recognising the extension in the name and reading the file: this file should only contain the sequence itself, with characters 'A', 'C', 'G', 'T' and '-' for mismatches.
Whenever a sequence is passed to the client and server from a .txt file, the client - upon receival of the output from the server and having displayed it – asks the user to input the name of a file where the sequence would be saved. Whenever a sequence is instead passed directly in the invoke command, the output is only displayed.
The script also handles a simple timer (via the built-in time python package) to measure the computation and response time.
Two options are provided: one using a very simple (i.e. BWT_server_simple.py) – yet not very fast or efficient – set of functions for conversion, and one more advanced using suffix arrays (i.e. BWT_server_advanced.py) – which is slightly more efficient. Both avoid using external packages that would be even more efficient, yet would miss the point of this project. Here's an example of the syntax required:
python3 BWT_server______.py
The dna_to_bwt() function used when the client invokes the bwt method adds '^' at the beginning of the sequence, and '|' at the end of it, to then perform the Burrows-Wheeler transformation and return the string with the markers included.
The bwt_to_dna() function used then the client invoked the ibwt method reorganises correctly the string using the '^' and '|' as markers, to then remove them before returning the string to the client.
In order to avoid hard-coding the server host:port in the script, those were written in a .json file which is accessed by the BWT_server.py script when invoked. Moreover, such host:port combination can be modified via the modify_config.py script and the following syntax:
python3 modify_config.py 0.0.0.0 5000
Since I decided to use flask instead of socket, there's no size cap for the file transmission from client to server. This could anyway be implemented via:
from flask import Flask app = Flask(name) app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024 # e.g., limit upload size to 16MB
Such a limitation could be required to avoid DoS (denial-of-services) attacks or resource exhaustion.
Moreover, since this tool is going to handle only DNA sequences and BWT strings, no EOF (end of file) terminator was used. And no multiprocessing approach was implemented - although both could be easily implemented without the need of huge alterations to the code.
Some examples of shell commands are provided in the commands_example.sh file, while some examples of DNA sequences and BWT strings are respectively in the dna_seq.txt and bwt_seq.txt files.
https://en.wikipedia.org/wiki/Burrows%E2%80%93Wheeler_transform https://www.cs.jhu.edu/~langmea/resources/bwt_fm.pdf https://www.cs.jhu.edu/~langmea/resources/burrows_wheeler.pdf