Histories of life: tree, web and ring models

[Jim]

Perspectives on the Phylogenetic Tree

Three of the models currently being considered in research on the history of life are called “tree,” “web” and “ring.”

Tree:

In tree models, all life on earth has one single common ancestor that lived almost four billion years ago. Two kinds of single-celled life forms evolved from that, Then multicellular life forms evolved from one or both of those.

 

[Jim]

Web:

In web models, some separate lines of ancestry go back to a diverse community of cells, instead of to a single organism.

 

[Jim]

Ring:

In ring models, the community of cells is drawn as a ring, instead of a band at the bottom of a web.

 

[Jim]

I want to discuss more in detail the other models that have been discussed in evolution research, besides tree models. First I’ll post what I understand about how tree models are constructed from looking at the genes in different species.

Imagine three species that I’ll call s1, s2 and s3. They all three have one gene in common, but s2 and s3 have another gene in common that s1 does not have. Imagine that if two species have a gene in common, it came from the same ancestor. Then all three species have a common ancestor, but s2 and s3 have an ancestor which was not an ancestor of s1. The tree would have a branch branching off into s1 and another branch, and that other branch would branch off into s2 and s3. Looking at more genes, more branches representing more species can be arranged in the form of a tree, like a family tree.

 

[Jim]

One of the reasons for discussions about other models besides tree models is because sometimes genes cross over from one species to another. That means that a tree constructed using some genes can be different from a tree using other ones. No single tree represents everything that the fossils tell us about the history of life. For some purposes one tree can be chosen and the others ignored, but not for all purposes. Some researchers have proposed what they sometimes call a “web” model, with separate lines of ancestry going all the way back to a time when there weren’t any lines of ancestry, and lines going between those representing genes crossing over between species. That could still be called a “tree,” considering that there are trees with many separate trunks.

 

[Jim]

Some researchers are promoting a model with separate lines of ancestry going all the way back to a self-perpetuating community of multitudes of protein-building teams of molecules, passing genes back and forth all across the community, more than to their own replicas if in fact they did replicate themselves individually. That means that there were no lines of ancestry, no living cells that could be ancestors of anything, In that model, eventually that community divided up into two or three groups that mostly only shared genes within each group. The first living cells that passed their genes on to their own replicas more than sharing them with each other, evolved separately within each group.

 

[Jim]

One reason that some researchers are discussing other models besides tree models is because tree models don’t show genes crossing over from one branch to another. That can be patched up by adding some horizontal lines, but the farther back we look, the more horizontal lines there would be, some of them crossing over some vertical lines to connect to other ones. Another problem is that there are some reasons for allowing the tree to have more than one stem or trunk. Then it looks more like a network or a web, so some researchers are questioning the value of persisting in calling it a “tree.” If it looks more like a network or a web, why not just call it that?

 

[Jim]

As I understand it, no one is denying that life could have started in a multitude of places, or even in the same place more than once. If so, then there might never have been anything that was an ancestor of all life on earth today. Possibly most researchers do think that only one of those was an ancestor of all life today.

 

[Jim]

I’m planning to discuss the history of classifying plants and animals, as part of the context for discussing trees and webs of life. I’ll be posting links in this thread as references for that discussion.

History of Plant Systematics

 

[Jim]

Plants have been classified in different ways for different purposes, for example according to their medicinal effects, or how much water or sunlight they need. For some purposes they have been classified according to their appearances. For example, the purpose might be to find a plant in a list according to its appearance and other features that are easily observed, to find information about it that is not so easy to see.

 

[Jim]

Biological classification decreases confusion and facilitates open dialogue in the scientific community, and scientists employ it to clearly identify species of flora, fauna or microorganisms. Without classification schemes, it would be nearly impossible for biological research to continue on plants, microorganisms and animals.

- What Is the Purpose of Classification Schemes in Biology?

With the influx of exotic species in the Age of Exploration ... most authors were far more interested in the medicinal properties of individual plants than an overarching classification system. ... Bauhin described over 6000 plants, which he arranged into 12 books and 72 sections based on a wide range of common characteristics. Cesalpino based his system on the structure of the organs of fructification ...

The book that had an enormous accelerating effect on the science of plant systematics was Species Plantarum (1753) by Linnaeus. It presented a complete list of the plant species then known to Europe,[1] ordered for the purpose of easy identification using the number and arrangement of the male and female sexual organs of the plants.

- History of plant systematics - Wikipedia

 

[Jim]

I’ve decided to post random thoughts here when they come to me, about models of the history of species.

- Adding a trunk at the bottom of a tree model with more than one trunk doesn’t change the amount of deviation from the data, if that trunk doesn’t correspond to any data.

- Tree construction processes include matching sequences to each other by adding, deleting and modifying parts of the sequences.

- A process that is designed to construct a tree will do so even if the data fed into it is totally random.

 

[Jim]

The accuracy of multiple sequence alignment programs has been the object of numerous comparative studies, which evaluate alignments either by using trusted reference alignments obtained from structural data, or by using simulation. Unfortunately, both approaches have flaws. ...proteins with resolved structure remain a small and highly biased sample of all proteins. In addition, homology inferred from structural information is inherently restricted to conserved regions, thereby providing little guidance for correct gap placement. The other approach to validating alignments is simulation. Yet, results obtained from simulated data strongly depend on the choice of model used to generate the data, and most biological processes are difficult to model realistically.

Phylogenetic assessment of alignments reveals neglected tree signal in gaps