MSAligner

A multiple sequence aligner written in python 🐍

Student project for the Python programming class of the Msc in Bioinformatics at the University of Paris Cité

MSAligner is a simplified re-implementation of Clustal and uses dynamic programming and a guide UPGMA tree to quickly align protein sequences.

Setting things up

Clone this repository in your local machine

git clone https://github.com/thaninacbn/MSAligner

Create a conda environment using the given .yml file

conda env create -f mseqalign.yml

And activate the environment

conda activate mseqalign

Getting started

Start by moving to the source code directory

cd src

Sample alignment

You can try MSAligner on sample fasta files provided in the /data directory.

ribosomes.fasta	calmodulins.fasta	wee1.fasta
50 AA	150 AA	600 AA
Homologs to human RL31 (accession number P62891)	Homologs to human CALM1 (accession number P0DP23)	Homologs to human WEE1 (accession number P30291)

To run MSAligner on ribosomes.fasta, run the following command:

python msaligner.py ../data/ribosomes.fasta output_ribosomes.fasta

To try it on the other files, simply run:

python msaligner.py ../data/calmodulins.fasta output_calmo.fasta

or

python msaligner.py ../data/wee1.fasta output_wee1.fasta

Run on your own sequences

To run MSAligner in default interactive mode with your own sequences, simply call from the command line:

python msaligner.py path/to/your/sequences.fasta output.txt

No matter the sequences you chose to use, a new /out directory will be automatically created and will contain the output.txt file.

Using MSAligner as a module

You can also use the MSAligner functions by importing it as a module in your own python scripts 🐍

To do so start by importing MSAligner:

import msaligner as msa

There are seven functions you can call from MSAligner:

read_fasta() and read_blosum(): read a fasta file and a blosum matrix respectively. Both return dictionaries

pairwise_alignment(): takes 2 sequences as an input, a blosum matrix and asks you if you wish to proceed to traceback (traceback="yes" or "no"). Performs the pairwise alignment of these two sequences according to the Needleman and Wunsch algorithm.

calculate_score(): creates a scores matrix for all the pairs of sequences in a fasta file by calling the pairwise_alignment() function

turn_scores_into_distance(): turns a scores matrix (contains negative and positive values) to a distance matrix (contains only positive values)

create_guide_tree(): creates a guide tree for multiple alignment using the UPGMA method

run_multiple_alignment() performs multiple sequence alignment on all proteic sequences from a dictionary of sequences using an UPGMA guide tree