Multiple Sequence Alignment Analysis

Primary Stack: Python • BioPython • Jupyter

Description

This project implements comprehensive multiple sequence alignment (MSA) analysis workflows, including consensus sequence calculation, position-specific scoring matrices (PSSM), sequence profile visualization, and Markov chain simulation for biological sequence analysis.

Problem

Analyzing biological sequences requires computing consensus sequences from multiple alignments, generating position-specific scoring matrices (PSSMs), and creating sequence logo visualizations. Additionally, understanding stochastic processes like Markov chains helps model biological phenomena such as state transitions.

Approach

The project implements three main components:

1. Consensus Sequence Calculation: Processes multiple sequence alignments (MSA) from FASTA files to derive consensus sequences using BioPython's AlignInfo module.

2. Sequence Profile Analysis: Generates position-specific scoring matrices (PSSMs) and converts them to probability-based profiles, then visualizes them as sequence logos using logomaker.

3. Markov Chain Simulation: Implements a discrete-time Markov chain simulator for modeling state transitions with custom transition matrices.

Tech Stack

• Python 3.x: Core programming language
• BioPython: Sequence alignment and analysis
• logomaker: Sequence logo visualization
• numpy: Numerical computations
• Jupyter Notebook: Interactive analysis (markov_simulation.ipynb)

Project Structure

msa-consensus-pssm-and-markov/
├── consensus_sequence.py           # Consensus sequence calculation from MSA
├── sequence_profile.py             # PSSM generation and sequence logo creation
├── markov_simulation.py            # Markov chain simulator
├── markov_simulation.ipynb         # Interactive Markov chain analysis
├── msa_output.fasta                # Multiple sequence alignment results
├── raw_sequences.fasta             # Input raw sequence data
├── requirements.txt                # Python dependencies
└── README.md                        # This file

Setup / Installation

1. Ensure you have Python 3.7+ installed:

python --version

2. Install required dependencies:

pip install -r requirements.txt

The requirements include:

• biopython>=1.79 - Sequence alignment and analysis
• logomaker - Sequence logo visualization
• numpy>=1.21.0 - Numerical operations
• matplotlib>=3.3.0 - Plotting support
• jupyter>=1.0.0 - Interactive notebook

Usage

1. Consensus Sequence Calculation

Calculate a consensus sequence from a multiple sequence alignment:

python consensus_sequence.py msa_output.fasta

Output: Prints the consensus sequence to stdout.

2. Sequence Profile Analysis

Generate PSSM, create probability-based profiles, and produce sequence logos:

python sequence_profile.py msa_output.fasta

Outputs:

• sequence_analysis_sequence_profile.csv - Position-specific probability matrix
• sequence_analysis_sequence_logo.png - Visual sequence logo (50x5 inches)

Custom Output Prefix: Set the MSA_OUTPUT_PREFIX environment variable to customize output filenames:

# Linux/Mac
export MSA_OUTPUT_PREFIX="my_analysis"
python sequence_profile.py msa_output.fasta

# Windows (Command Prompt)
set MSA_OUTPUT_PREFIX=my_analysis
python sequence_profile.py msa_output.fasta

# Windows (PowerShell)
$env:MSA_OUTPUT_PREFIX="my_analysis"
python sequence_profile.py msa_output.fasta

3. Markov Chain Simulation

Command-line version:

python markov_simulation.py

Interactive Jupyter notebook:

jupyter notebook markov_simulation.ipynb

The simulation models state transitions (e.g., city visits) using a transition matrix and tracks the frequency of each state over 100 steps.

Data Files

Input Files

• raw_sequences.fasta: Raw biological sequences in FASTA format
  • Contains unaligned sequences for initial processing

Output Files

• msa_output.fasta: Multiple sequence alignment results
  • Generated from MSA tool (e.g., Clustal Omega, MUSCLE)
  • Used as input for consensus and profile analysis

Key Features

Consensus Sequence Module

• Reads MSA files in FASTA format
• Uses BioPython's AlignInfo.SummaryInfo
• Applies dumb_consensus with threshold=0 for conservative consensus
• Handles single alignment files (typical for Clustal Omega output)

Sequence Profile Module

• Generates position-specific scoring matrices (PSSM)
• Converts raw counts to probability distributions
• Exports as CSV with sorted amino acid headers
• Creates high-resolution sequence logos (chemistry color scheme)
• Configurable for different sequence types

Markov Chain Simulator

• Discrete-time Markov chain implementation
• Customizable transition matrices
• Simulates N steps starting from any state
• Tracks and reports state frequencies
• Example: Models traveler movement between cities

Reproducibility Notes

Consensus Sequence

• Threshold parameter set to 0 for maximum conservation
• Alternative: Adjust threshold in dumb_consensus() for different stringency

Sequence Logos

• Generated with logomaker chemistry color scheme
• Figure size: 50x5 inches for high detail
• Stack order: small_on_top for visual clarity

Markov Simulation

• Uses numpy.random.choice with transition probabilities
• Set random seed for reproducible results:

  np.random.seed(42)

Troubleshooting

Import Errors:

pip install biopython logomaker numpy matplotlib jupyter

BioPython Version: Verify installation:

python -c "import Bio; print(Bio.__version__)"

Requires BioPython >=1.79

File Not Found: Ensure FASTA files are in the project directory or provide full paths:

python consensus_sequence.py /path/to/alignment.fasta

Logomaker Missing: The logomaker package is required for sequence logo generation:

pip install logomaker

Output File Customization: By default, output files are named sequence_analysis_sequence_profile.csv and sequence_analysis_sequence_logo.png. To customize the prefix, set the MSA_OUTPUT_PREFIX environment variable before running:

export MSA_OUTPUT_PREFIX="my_project"  # Linux/Mac
set MSA_OUTPUT_PREFIX=my_project       # Windows CMD

Expected Outputs

consensus_sequence.py

Consensus sequence printed to stdout
Example: MSFKILVACGLLL...

sequence_profile.py

Files created:
- sequence_analysis_sequence_profile.csv (CSV with probability matrix)
- sequence_analysis_sequence_logo.png (Sequence logo visualization)

markov_simulation.py / .ipynb

total number of cities visited: 100
Time in each city visited:
Berlin: 32
Munich: 38
Hamburg: 30