In the years before DNA sequencing scientists relied on morphology in order to determine the relationships of animals to one another. Tracking changes of anatomy in the fossil record as well as surveying the anatomy of living organisms was the best they could do. The discovery of DNA and the technology to cheaply sequence it helped scientists to resolve relationships that were ambiguous by looking at morphology alone. No method is foolproof, however, and the comparison of gene sequences has its own set of limitations and complexities. For example, point mutations can occur independently or revert to the original state, potentially obscuring true relationships.
A recent paper in Proceedings of the National Academy of Sciences provides a good example of an approach to resolving relationships that has been developed over the last 15 years or so. The central players are mobile genetic elements. These are pieces of DNA that, through various methods, can be copied and inserted into another part of the genome. These insertions represent unique events that are unlikely to occur independently, and the ancestral state can be confidently assumed to be the lack of the insertion. Thus, two genomes sharing the insertion are very likely to be related by descent, and any genome not containing the insertion (with no evidence for a deletion) represents a branching off before the insertion event occured. This should be clearer in a moment.
The authors of this paper used a family of long interspersed elements (LINEs) called L1 to investigate mammalian relationships. L1s are like retroviruses in that they use reverse transcription to insert into the genome. They only contain two or three gene genes which are used for their propegation, however they are not viruses–they remain in their cell of origin. L1s constitute 16% of the human genome and are present in a wide range of mammals. Here, the authors looked for L1s that are present in the same genomic location in a variety of animals. The following figure summarizes their findings.
This phylogeny is read similar to a genealogical chart. Moving left to right, each “T” intersection represents a split and if L1s are indicated, then all animals encompassed at that point share those L1s in the same genomic location. (Note: that doesn’t mean on the same chromosome, it means in the equivalent gene context.) The INT labels designate a particular L1. We can see that everything from bats to rabbits share 10 L1s (INT1068, INT1098, etc.), but only animal groups represented by bats, horses, and dogs and cats share four certain L1s (INT165, INT265, etc). It doesn’t take much time to see that there is a nested hierarchy. That two groups of animals would independently have an L1 insert into the same place is possible but unlikely; that it would happen several times over is even more unlikely. Thus the sharing of L1s–especially multiple L1s–is a good indicator of common descent.
This paper is also good because it presents an anomaly: INT283 makes it look like dogs and horses are more closely related to cows and whales than to bats. The authors rule out an independent insertion event, so how can this be? The answer that the authors favor is called incomplete lineage sorting. Each L1 insertion examined here occured in an individual some time in the distant past and spread through the population until it became fixed (ie. not polymorphic–or in other words–present in all individuals). If the L1 was neither harmful nor helpful (ie. neutral), it might take millions of years to become fixed. If the time to fixation encompassed speciation events, it is possible that the L1 would not be present (or would die out) in some lineages but ultimately become fixed in others, thus giving an incongruent picture. So what probably happened is that an ancestor of cows, dogs, horses, and bats had an L1 insertion that spread into the population. Some members of the population that split off and ultimately gave rise to cows and whales had the insertion and it became fixed. The same holds for the groups that gave rise to dogs or horses. However the L1 insertion failed to become fixed in the ancestors of bats and died out.
Similar studies have been done to resolve closer relationships. For example a type of mobile element that only occurs in primates has been used to help solidify primate relationships and it should come as no suprise that humans remain most closely related to chimpanzees. Because of their abundance and known mechanism of spread, mobile genetic elements such as L1s are powerful tools for determining evolutionary relationships. They also seem to strongly cut against the argument that similarity in DNA sequence between animal groups is merely a reflection of God’s economy of design.
Nishihara, Hasegawa, and Okada.
Pegasoferae, an unexpected mammalian clade revealed by tracking ancient retroposon insertions. PNAS vol. 103 no. 26 pp.9929-9934. (
This article is freely available here.)