Day 9: Molecular Phylogenetics: Tree Construction, Evolutionary Models & Bootstrapping

1. Evolutionary Foundations of Molecular Phylogenetics

Molecular phylogenetics reconstructs the historical evolutionary relationships among biological entities (genes, genomes, or species) through comparative mathematical analysis of aligned nucleotide or protein sequences. A phylogenetic tree is a formal graph composed of:

• External Nodes (Leaves/Taxa): Observed contemporary biological sequences (Operational Taxonomic Units, OTUs).
• Internal Nodes: Hypothetical common ancestors inferred through divergence points.
• Branches: Lineages connecting nodes. In phylograms, branch lengths are quantitatively proportional to the amount of inferred evolutionary change (e.g. substitutions per site).
• Roots: The deepest internal node representing the most recent common ancestor of all taxa in the tree (established using an outgroup).

2. Methodological Taxonomy of Tree Reconstruction

3. Nucleotide Substitution Models

Raw sequence distances underestimate true evolutionary divergence because multiple mutations frequently occur at the same nucleotide site over millions of years (homoplasy, transitions, transversions, back-mutations). Mathematical substitution models correct for these hidden mutations:

4. Statistical Tree Assessment: Felsenstein Non-Parametric Bootstrapping

To assess how confident we are in individual internal nodes/clades, Joe Felsenstein introduced non-parametric bootstrapping:

1. The original alignment of length $L$ is randomly resampled with replacement to generate a pseudo-replicate alignment of length $L$.
2. A tree is reconstructed from this pseudo-replicate.
3. Steps 1 and 2 are repeated 1,000 times.
4. The Bootstrap Value on an internal node represents the percentage of replicate trees (e.g. $98\%$) in which that exact bipartition of taxa was recovered. Nodes with bootstrap support $\ge 70-80\%$ are considered statistically well-supported.

Production Python Script: Tree Building & Newick Visualization with Bio.Phylo

#!/usr/bin/env python3
"""
Day 9 Practical: Distance Matrix Calculation & Neighbor-Joining Tree Construction
"""

from Bio import AlignIO
from Bio.Phylo.TreeConstruction import DistanceCalculator, DistanceTreeConstructor
from Bio import Phylo
import sys

def build_nj_phylogenetic_tree(msa_file: str):
    """Compute distance matrix and construct Neighbor-Joining phylogenetic tree."""
    print(f"Loading alignment from {msa_file}...")
    alignment = AlignIO.read(msa_file, "fasta")
    
    # 1. Compute pairwise identity distance matrix
    calculator = DistanceCalculator("identity")
    distance_matrix = calculator.get_distance(alignment)
    print("
Pairwise Evolutionary Distance Matrix:")
    print(distance_matrix)
    
    # 2. Construct Neighbor-Joining Tree
    constructor = DistanceTreeConstructor(calculator, "nj")
    tree = constructor.build_tree(alignment)
    tree.rooted = True
    
    # 3. Print ASCII phylogram to terminal
    print("
Inferred Neighbor-Joining Phylogenetic Tree:")
    Phylo.draw_ascii(tree)
    
    # 4. Save Newick format
    Phylo.write(tree, "inferred_tree.nwk", "newick")
    print("Saved Newick tree to inferred_tree.nwk (compatible with FigTree / iTOL).")

if __name__ == "__main__":
    # Test on insulin homologues from Day 8
    build_nj_phylogenetic_tree("insulin_homologues.fasta")