Molecular biology and comparative anatomy
Comparative anatomy: a recap and two additions
-
Homologous structures — same underlying anatomy, possibly different function — are evidence of divergent evolution and common ancestry.
-
Analogous structures — same function, different underlying anatomy — are evidence of convergent evolution and similar selection, not relatedness.
-
Two further kinds of anatomical evidence complete the picture:
-
Vestigial structures are reduced, non-functional or greatly reduced remnants of structures that were functional in an ancestor.
- Examples: the pelvic girdle of whales, which have no hind limbs; reduced wing bones in flightless birds such as the kiwi.
- They are evidence of ancestry because there is no reason for a structure to exist in reduced form unless it was inherited from an ancestor in which it worked.
- They persist because selection against a structure that costs little is weak, so reduction is gradual rather than immediate.
-
Comparative embryology — related species often show similar developmental stages even where the adults differ greatly.
- This is stronger evidence than adult anatomy, because developmental pathways are more conservative than adult form. A change early in development affects everything downstream, so early stages are strongly constrained.
Molecular evidence: proteins and DNA
-
All organisms use the same genetic code and the same basic biochemistry — DNA and RNA, the same 20 amino acids, ATP as the energy currency.
-
This universality is itself powerful evidence of common ancestry: there is no chemical necessity for the code to be the one it is, so its being shared by all life indicates inheritance from a common ancestor.
-
Protein analysis. Comparing the amino acid sequence of the same protein across species shows how similar they are.
- Fewer differences indicates a more recent common ancestor.
- Cytochrome c, a respiratory protein present in nearly all aerobic organisms, is the classic comparison.
-
DNA analysis. Comparing base sequences is more precise than protein comparison, because:
- The genetic code is degenerate, so some base changes do not change the amino acid. These silent changes are invisible in a protein comparison but visible in DNA.
- DNA comparison can use non-coding regions, which are under weaker selection and so accumulate differences more steadily.
- Whole genomes can be compared, giving vastly more data points than one protein.
Why molecular evidence is so powerful
-
It is quantitative. Anatomical similarity is a judgement; sequence difference is a number, so relationships can be compared and ranked objectively.
-
It is independent of the trait being explained. This is the decisive advantage. Convergence can make anatomy misleading, because strong selection produces similar forms in unrelated lineages. But sequences from elsewhere in the genome are not affected by selection on that trait, so they assess relatedness without circularity.
-
It works where anatomy cannot — between organisms with almost no comparable structures, such as bacteria and animals.
-
It provides a rough timescale. If mutations accumulate in a region at an approximately constant rate, the number of differences estimates the time since divergence — a molecular clock.
-
The molecular clock must be used carefully, and saying so gains credit:
- Rates differ between genes, between lineages, and between coding and non-coding regions.
- Clocks must be calibrated against independent evidence, usually fossils with known ages — so molecular dating is not independent of the fossil record after all.
Molecular evidence in New Zealand
- Tuatara were long assumed to be lizards on appearance. Anatomical and molecular evidence together place them in the Rhynchocephalia, a separate lineage from all lizards and snakes — an example of molecular data confirming that superficial similarity was misleading.
- Moa relationships were revised by DNA from subfossil bone. Moa turned out not to be most closely related to the geographically nearest ratites, and the ratite group as a whole is now understood to have lost flight more than once independently — a convergent outcome that anatomy alone had read as shared ancestry.
- Kākāpō have had their genome sequenced, and it is used directly in conservation to measure remaining variation and choose breeding pairs.
Selective advantage
- Molecular and anatomical comparisons are evidence, not responses. But they let selection be inferred:
- Highly conserved sequences — those differing very little between distantly related species — indicate strong stabilising selection, because almost any change was harmful and was removed. Cytochrome c is conserved for this reason.
- Rapidly changing sequences indicate either weak selection (so changes are neutral and accumulate freely) or strong directional selection, such as on immune genes under pressure from pathogens.
- Vestigial structures show selection relaxing: once a structure stops contributing to survival, alleles reducing investment in it are favoured because the resources are available for something else.
Worked Example
Worked Example
Four species are compared. The table shows the number of amino acid differences in the same protein, and the number of base differences in the same gene.
| Comparison | Amino acid differences | DNA base differences |
|---|---|---|
| Species A vs B | 2 | 14 |
| Species A vs C | 2 | 31 |
| Species A vs D | 19 | 96 |
- Species A and C look very similar and live in similar habitats on different continents.
- Species A and B look quite different from each other.
- A fossil with features shared by A and B is dated at 12 million years.
Explain what these data show about the relationships, and evaluate the conclusion.
Answer:
What the protein data alone would suggest. A–B and A–C both show 2 amino acid differences, so protein comparison alone cannot distinguish them — it suggests A is equally related to B and to C.
What the DNA data show. The DNA is far more informative:
- A vs B — 14 base differences
- A vs C — 31 base differences
So A is more closely related to B than to C, despite the identical protein result and despite A and C looking similar.
Why DNA distinguishes them when protein does not. The genetic code is degenerate — several codons specify the same amino acid — so many base substitutions are silent and do not change the protein. These silent changes are invisible to protein comparison but visible in DNA. DNA comparison therefore detects differences that have accumulated without altering the protein, giving much finer resolution.
Why A and C look similar despite being less closely related. This is convergent evolution. A and C live in similar habitats on different continents, so they have experienced similar selection pressures and independently evolved similar features. Their similarity is analogous, reflecting environment rather than ancestry.
This is exactly the situation in which molecular evidence is decisive. Anatomical comparison is misled here because the traits being compared are the very traits selection has shaped convergently. The DNA assesses relatedness using sequence independent of those traits, so convergence cannot distort it — which breaks the circularity that anatomy alone cannot escape.
Species D differs far more from A in both measures (19 amino acids, 96 bases), so it is the most distantly related of the four, having diverged from A's lineage earliest.
Constructing a relationship. From the DNA distances: A and B are most closely related, C branches off earlier, and D earliest of all.
Using the fossil to add a timescale. The fossil sharing features of A and B, dated at 12 million years, provides a calibration point. If 14 base differences correspond to roughly 12 million years of divergence, a rough rate can be estimated and applied to the other comparisons — this is a molecular clock.
Evaluating the conclusion.
The relationship conclusion is well supported, because the DNA evidence is quantitative, based on many sites, and independent of the convergent traits. The disagreement between the anatomical impression and the molecular data is itself informative — it identifies convergence rather than casting doubt on the method.
But three cautions apply to the dating:
- The clock assumes a roughly constant rate, and rates vary between genes, between lineages and between coding and non-coding regions. A rate estimated from one comparison may not apply to the others.
- The calibration depends on the fossil. The fossil's date, and the assumption that it sits on the A–B lineage rather than being a side branch, both carry uncertainty. So the molecular date is not independent of the fossil evidence — it inherits its uncertainty.
- Multiple substitutions at one site are undercounted. Over long periods a site may change more than once, or change back, so raw differences underestimate true divergence, and increasingly so for distant comparisons like A–D.
How to strengthen it. Compare more genes and ideally whole genomes, since a conclusion resting on many independent regions is far more secure than one from a single gene, and use several independently dated fossils as calibration points rather than one.