Annexin A11 Conformational Analysis

Undergraduate Research Project - University of São Paulo (USP)

A computational study investigating the conformational profile of Annexin A11 protein, comparing wild-type (WT) and mutant variants to understand the structural impact of disease-associated mutations.

---

📋 Project Overview

Annexin A11 is a calcium-dependent phospholipid-binding protein involved in various cellular processes. Mutations in ANXA11 have been linked to amyotrophic lateral sclerosis (ALS) and other neurological disorders. This project aims to characterize the conformational changes induced by specific mutations using molecular dynamics simulations and structural analysis.

Mutation Under Study

• P36R — Proline to Arginine substitution at position 36 (N-terminal region)

---

🧬 Data Source

Conformational ensembles were generated using BioEmu (Biological Ensemble Modeling Utility), which produces diverse protein conformations stored in .npz format containing:

• pos: Atomic coordinates (Cα positions)
• sequence: Amino acid sequence

---

📁 Project Structure

protein_analysis/
├── README.md
├── annexin_analysis/              # Main analysis package
│   ├── __init__.py               # Package exports
│   ├── config.py                 # Configuration and constants
│   ├── trajectory.py             # Trajectory loading and preprocessing
│   ├── analysis.py               # RMSF, PCA, convergence analysis
│   ├── contact_map.py            # Contact map computation
│   ├── visualization.py          # Plotting functions
│   └── utils.py                  # Utility classes (NPZ converter, etc.)
│
├── scripts/                       # Ready-to-run analysis scripts
│   ├── analyze_wt_rmsf.py        # RMSF analysis for WT
│   ├── analyze_wt_pca.py         # PCA analysis for WT
│   ├── compare_rmsf_wt_mutant.py # RMSF comparison WT vs P36R
│   ├── compare_pca_wt_mutant.py  # PCA comparison WT vs P36R
│   ├── compare_contact_maps.py   # Contact map comparison
│   ├── analyze_convergence.py    # Convergence analysis
│   ├── analyze_local_pca.py      # Local/regional PCA analysis
│   ├── align_trajectories.py     # Trajectory alignment utility
│   ├── convert_npz_to_pdb.py     # NPZ to PDB converter
│   └── run_full_analysis.py      # Complete analysis pipeline
│
├── results/                       # Output directory for figures
│
└── 1000_samples/                  # Trajectory data
    └── MutationConformationsBioEmu/
        ├── out_native/            # Wild-type conformations
        │   ├── samples.xtc
        │   └── topology.pdb
        └── out_mutant_P36R/       # P36R mutant conformations
            ├── samples.xtc
            └── topology.pdb

---

🔬 Analysis Methods

Analysis	Description	Script
RMSF	Root Mean Square Fluctuation per residue	analyze_wt_rmsf.py
PCA	Principal Component Analysis of conformations	analyze_wt_pca.py
RMSF Comparison	Compare flexibility between WT and mutant	compare_rmsf_wt_mutant.py
PCA Comparison	Compare conformational space sampling	compare_pca_wt_mutant.py
Contact Maps	Inter-residue distance analysis	compare_contact_maps.py
Convergence	Assess ensemble sampling completeness	analyze_convergence.py
Local PCA	Focused analysis on N-terminal region	analyze_local_pca.py

---

🛠️ Dependencies

pip install numpy mdtraj matplotlib seaborn scikit-learn scipy

Package	Purpose
numpy	Numerical computations, NPZ file handling
mdtraj	Molecular dynamics trajectory analysis
matplotlib	Visualization and plotting
seaborn	Statistical data visualization
scikit-learn	PCA and machine learning
scipy	Scientific computing

---

🚀 Quick Start

Running Individual Analyses

# Navigate to the project directory
cd protein_analysis

# Run RMSF analysis for wild-type
python scripts/analyze_wt_rmsf.py

# Compare WT vs P36R mutant RMSF
python scripts/compare_rmsf_wt_mutant.py

# Compare PCA projections
python scripts/compare_pca_wt_mutant.py

# Compare contact maps
python scripts/compare_contact_maps.py

Running Complete Analysis Pipeline

python scripts/run_full_analysis.py

This runs all analyses and generates a summary report in results/.

---

📦 Package Usage

The annexin_analysis package can be imported for custom analyses:

from annexin_analysis import (
    AnnexinConfig,
    TrajectoryLoader,
    ConformationalAnalyzer,
    ContactMapAnalyzer,
    ConformationalVisualizer
)

# Initialize with default configuration
config = AnnexinConfig()
loader = TrajectoryLoader(config)
analyzer = ConformationalAnalyzer(config)

# Load and process wild-type
wt = config.wt_variant
processed = loader.process_variant(wt)

# Compute RMSF
rmsf_result = analyzer.compute_rmsf(processed)
print(f"Mean RMSF: {rmsf_result.mean_rmsf:.4f} Å")

# Visualize
visualizer = ConformationalVisualizer(config)
visualizer.plot_rmsf(rmsf_result, filename="my_rmsf_plot.png")

Adding New Variants

from annexin_analysis import AnnexinConfig

config = AnnexinConfig()

# Create a new variant configuration
new_variant = config.create_variant(
    name="g154r",
    label="Mutant G154R",
    folder="out_mutant_G154R",
    color_index=2
)

---

🏗️ Architecture

The package follows object-oriented design principles:

Core Classes

Class	Description
AnnexinConfig	Centralized configuration for paths and parameters
TrajectoryLoader	Load, preprocess, and align MD trajectories
ConformationalAnalyzer	RMSF, PCA, and convergence analysis
ComparativeAnalyzer	Compare multiple variants
ContactMapAnalyzer	Contact map computation and comparison
ConformationalVisualizer	Publication-ready plotting

Data Classes

Class	Description
ProcessedTrajectory	Container for processed trajectory data
RMSFResult	RMSF analysis results
PCAResult	PCA analysis results with fitted model
ConvergenceResult	Convergence analysis results
ContactMapResult	Contact map data
ContactComparisonResult	Contact comparison data

---

📁 Protein Information

• UniProt ID: P50995
• Protein: Annexin A11 (Human)
• Length: 505 residues
• Reference Structure: AlphaFold predicted structure (v6)

Domain Architecture

Region	Residues	Description
N-terminal	1-198	Intrinsically disordered, proline-rich
Annexin Core	199-505	Conserved Ca²⁺-binding domains

---

📊 Key Findings

The analysis demonstrates that the P36R mutation affects the conformational space of Annexin A11:

1. RMSF Changes: Altered flexibility profile in the N-terminal region
2. PCA Shifts: Different conformational sampling patterns
3. Contact Changes: Modified inter-residue interactions

---

📚 References

1. AlphaFold Protein Structure Database - https://alphafold.ebi.ac.uk/
2. BioEmu - Conformational ensemble generation tool
3. MDTraj - https://mdtraj.org/

---

👨‍🔬 Author

Calistu - Undergraduate Research
University of São Paulo (USP)
7th Semester - Scientific Initiation (IC) Project

---

📄 License

This project is for academic research purposes.

---

Last updated: February 2026