bioinformatics-e1a-analysis

E1A Protein Analysis

Bioinformatics project focused on the analysis of E1A protein sequences using multiple sequence alignments (MSA), physicochemical property calculations, disorder prediction, and data visualization.

Overview

This project analyzes a collection of E1A protein sequences from different organisms using Python and bioinformatics tools. The workflow includes sequence analysis, amino acid composition, hydrodynamic radius estimation, charge distribution analysis, protein disorder prediction, and visualization of the obtained results.

The analysis was originally developed as part of a Bioinformatics course and later adapted into a reproducible GitHub project.

---

Objectives

• Analyze E1A protein sequence variability.
• Calculate sequence identity between proteins.
• Determine amino acid composition.
• Estimate hydrodynamic radius.
• Evaluate charge distribution using the Kappa parameter.
• Predict intrinsically disordered regions using AIUPred.
• Generate graphical representations of the results.

---

Methods

Sequence Analysis

The project uses multiple sequence alignments (MSA) in FASTA format and performs:

• Protein length calculation
• Pairwise sequence identity analysis
• Amino acid frequency analysis
• Net charge calculation

Physicochemical Properties

For each sequence:

• Hydrodynamic radius estimation
• Charge distribution analysis
• Kappa parameter calculation using LocalCIDER

Disorder Prediction

Protein disorder prediction is performed using AIUPred, generating disorder scores for each residue and integrating the results back into the aligned sequences.

Visualization

The workflow generates scatter plots to explore relationships between:

• Kappa vs Hydrodynamic Radius
• Sequence Length vs Hydrodynamic Radius

---

Key Findings

• E1A proteins show variability in sequence length across species.
• Charge distribution patterns can be quantified using the Kappa parameter.
• Hydrodynamic radius is associated with protein length and composition.
• AIUPred identifies extensive intrinsically disordered regions within E1A proteins.

---

Example Visualizations

Kappa vs Hydrodynamic Radius

Sequence Length vs Hydrodynamic Radius

---

Project Structure

bioinformatics-e1a-analysis/
│
├── data/
│   ├── E1A_MSA.fasta
│   ├── E1A_MSA1.fasta
│   ├── E1A_MSA2.fasta
│   ├── E1A_MSA_sin_gaps1.fasta
│   ├── E1A_MSA_sin_gaps2.fasta
│   └── AIUPred-1.0.zip
│
├── notebooks/
│   └── e1a_protein_analysis.ipynb
│
├── results/
│   ├── csv/
│   ├── figures/
│   └── *.txt
│
├── requirements.txt
└── README.md

Directory Description

Directory	Description
data/	Input FASTA files
notebooks/	Main Jupyter Notebook containing the complete analysis workflow
results/csv/	Generated tables and processed results
results/figures/	Generated visualizations and plots
results/	AIUPred output files
requirements.txt	Python dependencies required to run the project
tools/aiupred/	AIUPred package used for disorder prediction

Technologies

• Python
• BioPython
• LocalCIDER
• AIUPred
• Matplotlib
• Jupyter Notebook

---

Installation

1. Clone the repository

git clone https://github.com/fran9300/bioinformatics-e1a-analysis.git
cd bioinformatics-e1a-analysis

2. Install dependencies

pip install -r requirements.txt

3. Launch Jupyter Notebook

jupyter notebook

Open:

notebooks/e1a_protein_analysis.ipynb

and run all cells sequentially.

---

Required Dependencies

The project requires the following Python packages:

biopython
localcider
matplotlib
torch

All dependencies can be installed automatically using:

pip install -r requirements.txt

---

Results

The workflow generates:

Sequence Analysis

• Protein length calculations
• Pairwise sequence identity calculations
• Amino acid composition statistics
• Average amino acid frequencies

Physicochemical Properties

• Net charge calculations
• Hydrodynamic radius estimation
• Kappa parameter analysis

Disorder Prediction

• Intrinsic disorder prediction using AIUPred
• Disorder score integration into aligned sequences
• Gap-aware disorder alignments

Visualizations

• Kappa vs Hydrodynamic Radius
• Sequence Length vs Hydrodynamic Radius

Generated outputs are stored in the results/ directory.

---

AIUPred

This project uses AIUPred v1.0 (Publication release) for protein disorder prediction.

tools/aiupred/AIUPred-1.0/

Official repository:

https://github.com/doszilab/AIUPred

---

Reproducibility

All analyses can be reproduced directly from the provided notebook using the files included in the repository.

The workflow includes:

1. Loading multiple sequence alignments (MSA)
2. Sequence identity analysis
3. Amino acid composition analysis
4. Hydrodynamic radius calculation
5. Kappa parameter calculation
6. AIUPred disorder prediction
7. Result generation and visualization

---

Author

Francisco Kin

Bioinformatics and Data Analytics student interested in:

• Bioinformatics
• Data Analysis
• Machine Learning
• Biological Sequence Analysis
• Healthcare and Life Sciences Applications