Tree Calibration

Tree reconstruction algorithms estimate the evolutionary relationships between taxa from genetic distances. The resulting branch lengths represent relative evolutionary change, typically measured as substitutions per site, rather than absolute geological time.

Tree calibration converts these relative branch lengths into approximate divergence times by using one or more external estimates of when particular evolutionary events occurred.

The Molecular Clock project implements a simple educational calibration algorithm using a single calibration point.

Why Calibration Is Needed

Reconstruction algorithms such as UPGMA and Neighbor Joining can estimate the shape of an evolutionary tree from sequence data, but they cannot determine when each divergence occurred.

For example, a reconstructed branch length of:

0.05 substitutions per site

indicates that approximately five substitutions have occurred for every hundred nucleotide positions, but it does not indicate whether those substitutions accumulated over:

• 500 thousand years
• 5 million years
• 50 million years

To estimate geological time, the tree must be calibrated using independent evidence.

Calibration Points

A calibration point associates one node in the reconstructed tree with a known or estimated age.

In this project, the calibration point is defined by:

• Two taxa
• Their most recent common ancestor (MRCA)
• An estimated divergence time expressed in millions of years

For example:

Taxa	MRCA age
Taxon A, Taxon B	20 million years

The calibration algorithm locates the MRCA of the specified taxa and determines its reconstructed depth within the tree.

Scaling the Tree

Suppose the reconstructed tree estimates that the calibration node lies:

0.08 substitutions per site

from the descendant taxa.

If that ancestor is known to have existed:

20 million years ago

then one unit of evolutionary distance corresponds to:

20 / 0.08 = 250 million years

This value is the global scaling factor.

Every branch length in the reconstructed tree is multiplied by this factor.

For example:

Original branch length	Calibrated branch length
0.02	5 million years
0.04	10 million years
0.08	20 million years

The tree is therefore converted from relative evolutionary distance into estimated geological time.

Algorithm

The calibration procedure is straightforward:

1. Parse the reconstructed Newick tree.
2. Locate the MRCA of the selected taxa.
3. Calculate the reconstructed depth of that node.
4. Calculate the scaling factor

scale factor = calibration age / reconstructed node depth

5. Multiply every branch length by the scaling factor.
6. Export the calibrated tree together with calibration metadata.

The topology of the tree is unchanged.

Only the branch length units are modified.

Example

Suppose a reconstructed tree contains:

((A:0.04,B:0.04):0.06,C:0.10);

If the MRCA of A and B is known to be 20 million years old, the reconstructed depth is:

0.04 + 0.04 = 0.08 substitutions per site

The scaling factor becomes:

20 / 0.08 = 250

The calibrated tree is therefore:

((A:10.0,B:10.0):15.0,C:25.0);

The branch lengths now represent estimated millions of years rather than substitutions per site.

Relationship to the Molecular Clock

Calibration relies on the molecular clock principle that genetic differences accumulate approximately in proportion to elapsed evolutionary time.

The calibration process estimates the conversion between:

• Substitutions per site
• Millions of years.

In this implementation a single global rate is assumed across the entire tree.

This is consistent with the assumptions made by UPGMA and provides a simple educational demonstration of molecular dating.

Limitations

This implementation is intentionally simplified.

It assumes:

• A single calibration point
• A single global evolutionary rate
• An exact calibration age
• No uncertainty in branch lengths

Real molecular dating methods are considerably more sophisticated. Modern software such as BEAST simultaneously estimates evolutionary rates and divergence times using multiple calibration constraints, statistical models and Bayesian inference.

Those methods are beyond the scope of this project.

Educational Purpose

The objective of this implementation is not to produce publication-quality divergence dates.

Instead, it illustrates the final step in the molecular clock workflow:

Together with the simulation, distance estimation and reconstruction tools, calibration completes the educational pipeline from DNA sequence evolution to an approximately dated phylogenetic tree.