More Recent Comments

Tuesday, December 09, 2014

How many microRNAs?

MicroRNAs are a special class of small functional RNA molecules. The functional RNA is only about 22 nucleotides long and most of the well-characterized examples bind to mRNA to inhibit translation and/or destabilize the message.

The big questions for many of us are how many different microRNAs are there in a typical cell and how many of them have a real biological function. These questions are, of course, part of the debate over junk DNA. Are there thousands and thousands of microRNA genes in a typical genome and does this mean that there's a lot less junk DNA than some of us claim?

The journal Cell Death and Differentiation has devoted a special issue to microRNAs [Special Issue on microRNAs – the smallest RNA regulators of gene expression]. There are four reviews on the subject but none of them address the big questions.

That didn't stop the journal from leading off with this introduction ...
It is now well recognised that the majority of non-protein-coding genomic DNA is not “junk” but specifies a range of regulatory RNA molecules which finely tune protein expression. This issue of CDD contains an editorial and 5 reviews on a particular class of these regulatory RNAs, the microRNAs (miRs) of around 22 nucleotides, and which exert their effects by binding to consensus sites in the 3'UTRs of mRNAs. The reviews cover the role of miRs from their early association with CLL to other forms of cancer, their importance in the development of the epidermis and their potential as disease biomarkers as secreted in exosomes.
I'm not certain what the editors mean when they say that "it is now well recognised ..." I interpret this to mean that there are a large number of scientists who are completely uniformed about the structure of genomes and the debate over junk DNA. In other words, it is now well recognized that some scientists don't know what they are talking about.

I don't know any expert who would claim that 50% of large genomes consist of genes that specify regulatory RNAs involved in fine-tuning protein expression. Do you?

On a related issue, Wilczynska and Bushell begin their review with ...
Since their discovery 20 years ago, miRNAs have attracted much attention from all areas of biology. These short (~22 nt) non-coding RNA molecules are highly conserved in evolution and are present in nearly all eukaryotes.
Sequence conservation is an important criterion in deciding whether something is functional. In order to use conservation as a measure of function you have to establish some standards that let you distinguish between sequences that are "conserved" by negative selection and those that have drifted apart by random genetic drift.

What do Wilczynska and Bushell mean when they say that microRNAs are "highly conserved"? The most highly conserved genes exhibit about 50% sequence identify between prokaryotes and eukaryotes. They are almost identical within mammals. Other highly conserved genes are about 80% identical within animals (e.g. between insects and mammals). As far as I know, the sequences of most putative microRNAs aren't even similar within mammals and certainly not between mammals and fish.

The phrase "highly conserved" has become meaningless. It's now a synonym for "conserved" because nobody ever wants to just say "conserved" and they certainly don't want to say "moderately conserved" or "weakly conserved" even if it's the truth.


78 comments :

Athel Cornish-Bowden said...

I can't wait to see what Dan Graur has to say about this. I don't suppose that he'll be as polite as you have been.

And yes, it's quite meaningless to use the term "highly conserved" without defining it.

bFast said...

Dr. Moran, what percentage of the human DNA do you think is functional? How much do you think is junk?

DGA said...

Read past entries on this site to find out. It would seem that about 8 %, from what I recall reading, is functional, or known to have a function, either protein coding or regulatory, and the rest is of unknown or no function, most likely no function

bFast said...

Dr. Moran, do you agree with DGA's figures?

Larry Moran said...

@bFast

About 90% of mammalian genomes is junk.

bFast said...

So what would your answer have been a decade ago?

Unknown said...

I think there are some good reasons to assume that identifiable microRNAs tend to be highly conserved and that is that you need a good chunk of the sequence to identify homologs. When you are looking at 20BPs 50% sequence identity wouldn't give you a decent BLAST hit. I wouldn't be surprised if 18+ BPs are conserved between say nematode and insect miRNAs (OK, that's cheating a bit - I recently had results to that effect personally communicated by somebody working on microRNAs). 80+% sequence identify between identifiable homologs doesn't seem far fetched as far as I'm aware (I'll talk to the person in question on friday, maybe I'm remembering things wrong).
The main issue with the editorial seems to be that it overestimates how much of the non-coding regions are functional. There are less than 600 known microRNAs+snRNAs and snoRNAs in Drosophila. Even with the size of the larger snRNAs, you end up with 60 kilobases in total, or 0.05% of the genome.

Georgi Marinov said...

It is now well recognised that the majority of non-protein-coding genomic DNA is not “junk” but specifies a range of regulatory RNA molecules which finely tune protein expression

This is of course false.

But conservation for miRNAs is very different from conservation for protein coding genes - there is the seed, which is critical for function, and then there is the rest of the sequence.

Anonymous said...

Doesn't it bother you a bit that the "evolu-prophets can't agree on anything...?
Evolution is supposed to be so simple... so smooth... and you... no doubt a very intelligent scientist... MAKE A DECISION to side with them... I just would like to know why...? You must have some secrets...

Anonymous said...

Simple? I'm not sure how anybody could conclude that a process dependent on innumerable complex and interlocking mechanisms, operating at a vast range of functional levels, should be simple to model. Could you clarify?

Larry Moran said...

bFast asks,

So what would your answer have been a decade ago?

About 90% of mammalian genomes is junk.

bFast said...

'Seems that the assumption used to be that about 98% was junk. Was that not the dominant view in the recent past?

apalazzo said...

Well there ARE miRNAs that are highly conserved ... Let7 for example ... although it is also probably the most important of all miRNAs. (Thought you would like to know)

Larry Moran said...

What's your definition of "highly conserved"?

Larry Moran said...

Nobody who was knowledgeable about the subject ever believed that all noncoding RNA was junk.

apalazzo said...

100% identical between human and worm (my deleted comment below was exactly this statement, but I wanted to "reply" not post a new comment - these blogs have really evolved since I left the blog-o-sphere;) )

Mikkel Rumraket Rasmussen said...

bFast, where do you get that 98% number from? Give a reference.

Tom Mueller said...

For what it's worth... apalazo's take on "highy conserved" has to be correct on this one.

Making hitherto impossible research now possible

eg.

https://www.sciencemag.org/content/334/6059/1091.abstract?related-urls=yes&legid=sci;334/6059/1091

Athel Cornish-Bowden said...

In my case I know exactly what I thought 10 years ago, because in 2004 I wrote it down: "In humans the junk DNA accounts for around 90% of all the DNA".

apalazzo said...

Let-7 is special though. It's deletion has the identical phenotype as deleting dicer or the main miRNA-associating ago in C. elegans. They die a catastrophic death where their insides burst out. They also show profound developmental defects.

In most metazoans Let-7 is also intricately involved in embryonic development and differentiation. Lin28 acts to inhibit Let-7 function, and it is one of the "magical" factors that you can express to generate mammalian iPS cells.

Larry Moran said...

Alex Palazzo says,

100% identical between human and worm

I found a paper [PDf] showing that the 21 bp let-7 sequence can be found in the genomes of humans and Drosophila. There's a perfect match in sequence identity.

The match is confined to bilaterian animals meaning that it has only been conserved for about 600 million years. There's no let-7 in sponges, plants, protists, or bacteria.

I'm still not sure this qualifies as "highly conserved."

The functionally related lin-4 sequence from C. elegans is not at all conserved. In fact, there are no related sequences in Drosophila or humans.

Let's not lose site of the main point. Is is correct to say that microRNAs, in general, are "highly conserved"?

Tom Mueller said...

Hi Larry

Kevin Petersen would answer a resounding “YES” to your question:

http://mbe.oxfordjournals.org/content/early/2013/08/02/molbev.mst133
http://onlinelibrary.wiley.com/doi/10.1002/bies.201200055/abstract
http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199549429.001.0001/acprof-9780199549429-chapter-15

The last link in particular is most worthy of perusal (IMHO)

Larry Moran said...

Peterson argues that microRNAs make good phylogenetic markers because they are often restricted to particular clades. For example, many of the Drosophila microRNAs are only found in Drosophila and closely related species. Mosquitos and their relatives have their own distinctive, and diagnostic, microRNAs.

Cnidarians have a distinctive family of microRNAs that aren't found anywhere else in the animal kingdom. It's true that within a distinctive clade, the sequence identity is often 21/21 or 20/21. That's one measure of "highly conserved" but the fact that these "highly conserved" sequences are restricted to smallish clades is not consistent with other definitions of "highly conserved."

It would be like referring to an enzyme that's only found in primates and nowhere else. The active site residues are invariant. Is this a "highly conserved" protein?

Tom Mueller said...

Hi Larry,

Maybe I am reading too much into this paper:

http://sandwalk.blogspot.ca/2011/12/cambrian-conundrum-fossils-vs-genes.html

But I imagined Petersen tracking high conservation back to the Cambrian.

John Harshman said...

I think you have to make a distinction between the sequence being highly conserved (changes very little) and the microRNA itself being highly conserved (remains in taxa descended from the one in which it arose). If a microRNA arose fairly recently it would be difficult to tell, so only comparatively ancient ones are at issue. It seems clear that by the first definition, they're highly conserved. Since you haven't considered loss, but only presence, you haven't given us anything to work with for the second. If a microRNA arose in one of these "smallish clades" (and that's pushing a definition too since one of your examples was cnidarians), that doesn't tell us whether it's conserved, unless it was also lost several times within cnidarians. Now if it arose in the metazoan ancestor and is preserved only in cnidarians, you might have something. But good luck figuring that out.

Larry Moran said...

@Tom

Go to this post: Do Invertebrates Really Make Up 80% of All Species on Earth?. Look at the phylogenetic tree. Do you see "Animals" on a little twig in the lower-right corner?

That's where all the animal microRNAs would be found if they arose in the animal lineage during the Cambrian. Do you think that a gene that was confined to that little twig deserves to be called "highly conserved"?

Georgi Marinov said...

Come on, you know that eukaryotes are equivalent to metazoans + fungi + plants, and nothing else is worth bothering with.

Anonymous said...

Well, yes, I would consider the Let-7 mRNA highly conserved if it's conserved among most animals and goes back 600 million years. But then I am pretty strongly biased toward multicellular eukaryotes. : )

judmarc said...

Georgi Marinov, I was just reading Koonin's The Logic of Chance and chuckling over the following:

"[E]ukaryotic life is mainly defined by the enormous diversity of unicellular forms, whereas the conspicuous, large multicellular creatures are only offshoots in the three eukaryote branches founded by protists."

Georgi Marinov said...

Great book.

And nothing to disagree with in that quote. A commonly given figures is between 80 and 100 separate eukaryotic lineages that are at least as deeply branching as the three well-known ones (though at least the red and brown algae should also be a lot more popular than they currently are), metagenomics has revealed a number of other candidates, and who knows what else might be lurking out there with weird rRNA sequences that are not captured well with existing primers...

Tom Mueller said...

LOL with Georgi

Come on, you know that eukaryotes are equivalent to metazoans + fungi + plants, and nothing else is worth bothering with.

Georgi, you forgot Protista. ;-)

... though as I post, I notice you and judmarc already address this omission

But you both do raise and interesting point. Being a further step removed from reality (i.e a High School teacher with hopelessly outdated resources) I am in desperate straits.

In my last Grade 11 test, I provided students a simplified phylogenetic tree of Archaea and Bacteria generating 9 “groups” of Eukarya. Some of those Eukarya “groups” were multicellular, some single celled and some simultaneously both. I asked students to draw and label a big circle around the groups traditionally considered “Protist” and then answer two questions.

Why will “Protist” as a taxon soon go out of date no differently than the outdated taxon “Monera”?

Why will the taxon “Kingdom” similarly be outdated very soon?


All of this of course is a direct contradiction of everything written in the “textbook”.

Why do I bring this up? I need to impress upon those with influence in higher circles that high school texts are in desperate need of remediation and correction!

Could any and all please get the message out to curriculum committees and textbook publishers?

Tom Mueller said...

@ Larry & @ John

Thank you for clearing up some fuzzy thinking on my part.
When I was thinking “highly conserved” I was thinking along the lines of John’s first distinction above…

1 - … a distinction between the sequence being highly conserved (changes very little)

... as opposed to

2 - … and the microRNA itself being highly conserved (remains in taxa descended from the one in which it arose).


According to Kevin Petersen in the source I cited above

http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199549429.001.0001/acprof-9780199549429-chapter-15

1) new microRNA families are continually being incorporated into metazoan genomes through time; 2) they show very low homoplasy, with only rare instances of secondary loss, and only rare instances of substitutions occurring in the mature gene sequence; and 3) are almost impossible to evolve convergently.

… making miRNA useful in constructing phylogenetic trees. Because they tend to be very (for lack of a better word) “immutable” once established (Petersen’s points #2 & #3) explaining my (perhaps naïve) support for Alex Palazzo’s contention.

I think you are focusing on Kevin Petersen’s point #1 above which to my reading corresponds with John Harshman’s point #2

I hope I am getting this all right and have not made a hash of everything.

Larry Moran said...

Barbara says,

Well, yes, I would consider the Let-7 mRNA highly conserved if it's conserved among most animals and goes back 600 million years. But then I am pretty strongly biased toward multicellular eukaryotes.

Let's explore that bias. :-)

Imagine that there is a protein that's only found in cyanobacteria and in no other prokaryotes and no eukaryotes. Imagine that the sequence of that protein is 80% identical in all cyanobacteria.

Is it highly conserved?

What if it's only 60% identical?

What if the protein/gene is confined to a subclass of cyanobacteria, say Nostocales, that diverged from a common ancestror more than 600 million years ago? Does that still count as highly conserved?

John Harshman said...

Does that still count as highly conserved?

I don't know if 60% should be considered highly conserved, but of course "highly" has an arbitrary cutoff. Being found only in a restricted clade doesn't seem to me to be any sort of barrier to high conservation, though. Why focus on that?

John Harshman said...

Larry: what does conservation have to do with universality? I think you are confusing two completely separate issues.

Unknown said...

You are getting off track by moving to coding sequences here. miRNAs are non-coding and as Georgi has pointed out that leads to a difference in conservation - you simply don't have synonymous sites. You also ignore my comment on detecting homologs given very short sequences. Any two random sequences of length 22 will match in at least 11 positions with a probability of 0.46%. Given the size of genomes you could never tell a homolog from random matches. Hence detectable miRNAs tend to be the ones with high sequence identity. We are talking about the subset of detectable miRNAs in at least bilaterians and these are highly conserved by any reasonable definition. If you look at this in terms of BPas (i.e. bases times total time in the phylogeny), you end up with estimates for substitution rates that are a couple of orders below the neutral rate.
Since that's a criterion that is conceptually similar to what we do with dS/dN ratios for coding regions I'd consider it meaningful. It's also not specific to any particular clade. If substitution rates for some sequence are below neutral rates by that much in any clade that sequence is highly conserved.

Now, my hunch is that there isn't anything special about miRNAs and that there are a lot far less conserved ones. But detecting these is tough. You want transcriptomic data, preferably sampling various tissue types in various states of development for a wide range of organisms. And library preparation is somewhat different if you are specifically looking for miRNAs. For the same reason I'd be wary in discussing differences in miRNA repertoires right now - I don't think we do have a deep enough sampling to be able to conclude that some miRNAs are absent in some taxa, rather than just not highly enough conserved between two taxa to be detectable as homologs.

Finally, I'd just like to point out that it seems remarkably silly to hold a position that would exclude any relatively novel genes (and with these being ~20% of the age of life, the term novel seems quite off to me) from being highly conserved while at the same time holding the eminently reasonable position that the degree to which a sequence is conserved is informative on whether it is functional.

Anonymous said...

Re: Protista.

I am reminded of a discussion some years ago. Prof points out that Protista has to be broken up because it isn't a clade. I argue no. He points out that oomycota (protists) are more closely related to animals than to fungi or other protists and says, "So you have a problem." I respond, "No, you have a problem because you think Protista should be a clade. It's actually a group of taxa that are so different we'd treat them as kingdoms except that they aren't as diverse as animals, plants, and fungi." (I should have added, because most of them are very small.) This is so far outside his logic that he's momentarily dumbfounded. He says, "Nobody treats Protista that way." I say "I've always taught it that way," and he has no response. I like to think that's because he sees I'm right, though perhaps more likely it's because he sees I'm hopeless.

John Harshman said...

I suspect they're more diverse than animals, plants, and fungi, but they're harder to find and describe. What does that do to your claim? Anyway, I suspect Larry's complaint isn't about "protists", a logical class with membership criteria (single-celled eukaryote); it's about Protista, a formal taxonomic group. By modern standards those groups have to be monophyletic. If you don't like that, you should complain to the systematists. You can complain to me, if you want.

Tom Mueller said...

Part of the problem is that some Protista are multi-cellular

Georgi Marinov said...

Tom MuellerThursday, December 11, 2014 8:39:00 AM
All of this of course is a direct contradiction of everything written in the “textbook”.

Why do I bring this up? I need to impress upon those with influence in higher circles that high school texts are in desperate need of remediation and correction!

Could any and all please get the message out to curriculum committees and textbook publishers?


I have been removed from high school for a decade now, and I haven't seen an american high school textbook anyway. What exactly is the situation right now?

How is systematics treated in current books? Are they still presenting the old kingdom divisions, and is there any hint of cladistic thinking in them? I doubt they are presenting the whole diversity, but how bad exactly is it?

Piotr Gąsiorowski said...

Well, if one insists that a protist must by definition be single-celled (or at least consist of a small number of cells), then things like kelp (30-50 metres long and capable of growing 50 cm per day) will not be protists. But what will they be? If one excludes all red algae, there will be lots of unicellular non-protists. If one excludes only multicellular red algae, then some red algae will be protists and others will be ... well, something else. And of course Rhodophyta are not the only group including both "protistan" and "non-protistant" taxa (not to mention borderline cases). So the class is just traditional, not really logical.

John Harshman said...

There are of course borderline cases. Is Volvox multicellular or just colonial? Usually we say the latter. But Tom, which multicellular Protista are you thinking of?

Georgi Marinov said...

Slime molds?

Larry Moran said...

John Harshman asks,

Larry: what does conservation have to do with universality?

I really don't know the answer to your question but I think there's a problem. Let's say you discover a gene that's only found in humans and chimps but it's almost identical in both species. Is that a "highly conserved" gene? If so, it means that every single gene must be highly conserved because all of them will be nearly identical in some pair of very closely related species.

I think you are confusing two completely separate issues.

I think the idea of conservation has to be linked in some way to the size of the clade where it's found. I'm not sure how to define that relationship but I do think there's a connection.

Part of the problem is that the word "conservation" means that the similarity is due to negative selection. This doesn't apply to closely related species since any stretch of the genome is bound to be very similar just because they share a recent common ancestor (e.g. any 22 bp sequences at the same locations in the human and chimp genomes).

Thus, you can't talk about "highly conserved" unless you are referring to distantly related species that should have diverged by random genetic drift.

Tom Mueller said...

I was thinking exactly along the lines as Georgi

The Phaeophyceae or brown algae are not monophyletic to "Plants"

neither are they Plants nor Animal

so they are tossed into the "catch-all" drawer "Protista"

My students recognize that current high school texts are hopelessly out of date.

Tom Mueller said...

I am a huge fan of slime molds!

I had some email exchanges with John Bonner who has a summer cottage not far from where I reside.

https://www.youtube.com/watch?v=bkVhLJLG7ug

I teach my students slime molds are perhaps the planet’s first farmers/shepherds.

Here are some questions I give my students:

Slime Molds ingest their food; that makes Slime Molds different than Fungi which do not ingest their food. However, Slime Molds are still similar to Fungi by secreting digestive enzymes.

Speculate why Slime Molds secrete digestive enzymes even though they normally ingest their prey? (Hint: some of these enzymes specifically break down carbohydrates such as cellulose and chitin)

Is human nutrition more similar to Fungi or to slime molds?

Why are slime molds sometimes considered "transitional" to animals?

You may want to read these two news reports first:


http://news.bbc.co.uk/2/hi/science/nature/944790.stm http://www.bbc.co.uk/news/science-environment-12227168

Tom Mueller said...

of course I meant to say

neither are they Fungus nor Animal

Tom Mueller said...

Georgi asks

I have been removed from high school for a decade now, and I haven't seen an american high school textbook anyway. What exactly is the situation right now?

How is systematics treated in current books? Are they still presenting the old kingdom divisions, and is there any hint of cladistic thinking in them? I doubt they are presenting the whole diversity, but how bad exactly is it?


You have no idea!!!

A picture tells a thousand words. Check out this “corrected” diagram from Ken Miller’s and Joe Levine’s textbook that I am legally obliged to use in class.

http://kvhs.nbed.nb.ca/gallant/biology/Cladogram_animalia.jpg

Actually I use a version of the diagram in my final exam asking students to identify as many errors as they can (where molluscs were the side-brand to segmented annelids and arthopods, I kid you not) .

Minimum Expectations:

Humans are not the apex of evolution - there is no scala natura

The diagram fails to distinguish Lophotrochozoans and Ecdysozoans (arthopods, mollusks and annelids are not monphyletic!!!!)

The common ancestor to Lophotrochozoans, Ecdysozoans and Deuterostomes was probably coelomate: therefore modern acoelomates and pseudocoelomates can be best described as more recent atavisms present in all three lineages.

With a tip of the hat to John Harshman ( I hope I did not make a mess of this)

I also want my students to understand that it is possible that Deuterostomes may resemble that last common ancestor more than either Lophotrochozoans or Ecdysozoans. For example, that ancestor may have had a dorsal and not a ventral nerve cord meaning humans could perhaps in some sense be considered less “derived”… and insects more “derived”.

There are some other mistakes… but these were my minimum expectations.

Feedback, correction and suggestions for improvement gratefully appreciated.

Georgi Marinov said...

This is embarrassing - it is exactly the same thing I was taught when I was a kid back in the 90s/early 00s, and I was reading old Soviet and Bulgarian university textbooks from the 70s because I had no access to more modern literature.

The Lophotrochozoa/Ecdysozoa rearrangement was a shock to me when I first saw it. But I have gotten over it, and there is no excuse for still teaching that :(

The authors should be ashamed of themselves

Tom Mueller said...

Welcome to my nightmare!

Which provides opportunity yet again to express my gratitude for everybody's patient and indulgent altruism.

Participation on this forum has made me a better teacher.

Thank you!

Tom Mueller said...

erratum... and of course I meant to say

I was thinking exactly along the lines as Piotr...

so sorry. I am ulti-tasking right now.

I have to sign off... family obligations require my undivided attention.

Unknown said...

Larry, it's not a hard problem.
"I think the idea of conservation has to be linked in some way to the size of the clade where it's found. I'm not sure how to define that relationship but I do think there's a connection."

Let's say you have 5 recent taxa A,B,C,D and E and a phylogeny that looks like this:
((A,B),C),(E,F)
You use a decent molecular clock approach to date the divergences and obtain an age for the nodes: AE,AC,AB and EF. Then the total time in the phylogeny is 2AE+AC+AB+EF
Multiply this time with the neutral substitution rate (estimated from a pseudogene for instance) and the number of nucleotides in the sequence gives you an expected number of substitutions in the phylogeny under a neutral model. What's more the number itself is Poisson distributed, and this means that you have a null model for statistical tests. Since out total time calculation depends on the number of terminal taxa and the timing of their divergence this does capture all the factors one would expect to go into this.
And again, that's really not all that different from testing for selection using dN/dS ratios.

John Harshman said...

Larry,

Let's say you discover a gene that's only found in humans and chimps but it's almost identical in both species. Is that a "highly conserved" gene?

The answer is, of course, that you don't know. Only time will tell.

I think the idea of conservation has to be linked in some way to the size of the clade where it's found. I'm not sure how to define that relationship but I do think there's a connection.

Nope. It has to be linked to the *age* of the clade in which it's found. The size is irrelevant.

Thus, you can't talk about "highly conserved" unless you are referring to distantly related species that should have diverged by random genetic drift.

For sure. So why are you bringing up the sizes of clades? That's irrelevant.

John Harshman said...

Who is it tossing brown and red algae into the same bin as Protista? I haven't seen such a classification.

Georgi Marinov said...

I have seen it many times but I can't find an example right now

Tom Mueller said...

That same Miller& Levine high school text for example.

John Harshman said...

That seems odd. By the usual 5-kingdom classification, brown and red algae are, I believe, plants. But-but-but, you sputter, that's polyphyly! Not a big matter of concern to those who came up with it, apparently.

Larry Moran said...

@John Harshman

For sure. So why are you bringing up the sizes of clades? That's irrelevant.

I was using "size" as a convenient way of expressing whether the common ancestor is recent or ancient. I was thinking of the length of the branches which indicates the depth as well as the total number of nodes in a typical tree. I think you knew that.

You and I both know that different phyla, for example, have differing numbers of species ranging from one to millions. Give me a little credit.

Piotr Gąsiorowski said...

See here, for example. The history of the terms Protista, Protictista, Protozoa, etc. is discussed at some length by Scamardella (1999).

As noted above, some red algae (for example) are unicellular and look like this. If unicellularity is treated as a diagnostic criterion, Rhodophyta should be split between two "kingdoms" (or, alternatively, be regarded as "ambiregnal", the way the ICN and the ICZN treat slime moulds). But then Protista reduces to a purely descriptive label (= unicellular eukaryotes). One could just as well define "Bipedalia" as any tetrapods that use two legs for walking (birds, pangolins, humans, and possibly miscellaneous others).

Tom Mueller said...

@ Larry & @ Larry

Let's say you discover a gene that's only found in humans and chimps but it's almost identical in both species. Is that a "highly conserved" gene?

Instead of Chimps & Humans, how about Lampreys and Humans?

...or how about Tunicates & Humans?

What if sequence homology is almost identical in these instances?

I think I must be missing something.

Tom Mueller said...

re judmarc's quote above:


"[E]ukaryotic life is mainly defined by the enormous diversity of unicellular forms, whereas the conspicuous, large multicellular creatures are only offshoots in the three eukaryote branches founded by protists."

Sadly, that is an exact description of how phylogeny is typically taught!

http://www2.estrellamountain.edu/faculty/farabee/biobk/kingdoms.gif

I for one would like to know what would constitute an example of a modern "fungus-like" Protist?!

The problem is that high school texts are hanging on to tradition:

(Domain)
Kingdom
Phylum
Class
Order
Family
Genus
Species

So when do we study red and brown algae? Hmmm... Not with Plants - so Protists it is!

I argue that the easiest fix is to abolish the taxon "Kingdom" and be done with it.



Unknown said...

@Larry:
It's not completely irrelevant. Let's say you have a sequence of length 22 in two species that diverged 60 million years ago and there is one difference. On the other hand, you have 4 species, the first of which diverged 40 million years ago from the others, one diverged 25 million years ago from the others, and the last two split 15 million years ago and in one of them there is one difference. Which sequence is less conserved? In both cases the total time in the phylogeny is 120 million years, in both cases one substitution occured. The degree to which they are conserved is the same.

Tom Mueller said...

Reality Check:

Not too long ago, Larry asked the question:

That curriculum looks pretty impressive to me. What's the point of teaching "AP" biology in Canada?
http://sandwalk.blogspot.ca/2014/04/core-concepts-in-genetics.html

The best answer I can muster is that I inflicted much error upon my students until I signed on for AP. I blush to admit that I also taught that … large multicellular creatures are only offshoots in the three eukaryote branches founded by protists

AP Biology imposes a quality control on teachers. AP teachers have to get with the program and stay current. For example, textbooks must be less than 10 years old and must be approved.

With that in mind – allow me to present the most recent edition of one approved text that still employs the term “Protist” but in much modified context than other texts such as the Miller & Levine version cited above.

http://highered.mheducation.com/sites/0073383074/student_view0/chapter29/image_powerpoint_for_students.html

AP Biology only endorses the highest quality university-level texts.

Typical high school texts (like Miller & Levine) are quite a different story all together.

Sadly, the Raven phylogeny would be very unfamiliar to the majority of high school teachers who continue to espouse naïve a 6-kingdom mythology as found in all typical high school texts.
http://www2.estrellamountain.edu/faculty/farabee/biobk/kingdoms.gif

Georgi Marinov said...

Needless to say, this must have devastating impact on the understanding of evolution. All of the examples you posted present a progressive view of evolution. And we know all the things that follow from that view :(

John Harshman said...

I was using "size" as a convenient way of expressing whether the common ancestor is recent or ancient. I was thinking of the length of the branches which indicates the depth as well as the total number of nodes in a typical tree. I think you knew that.

No, I didn't. That was a very poor way of expressing yourself, especially when I had previously made the distinction between age and size. If you want to be understood, say what you mean, not something else entirely.

And Simon Gunkel makes a fine point. The total time covered by all branches of a tree is what you need to assess when deciding on sequence conservation. And it does seem to me that 100 million years is more than enough branch length to distinguish "converged" from "highly conserved". 100 million years total branch length is more than enough branch length to distinguish "conserved" from "highly conserved", much less "evolving neutrally". So all this talk about phyla, and assertions that Cnidaria is a "small clade" seem to show a serious misunderstanding.

Larry Moran said...

High school students need to learn that there's a lot more to life than mushrooms, maple trees, and moose. They need to learn about bacteria and single-cell eukaryotes. If you want to refer to single-cell eukaryotes as "protists" then that's okay with me.

The important point is that students learn of their existence and their diversity. Quibbling about cladistics and the exact relationship of some multicellular algae is distracting and counter-productive for high school students.

If you want to jump into that morass then you'd be better off discussing the Three Domain Hypothesis and whether archaebacteria really represent a distinct domain of life that gave rise to eukayotes. Or, for that matter, whether humans are a form of fish.

Tom Mueller said...

I have a naïve question tangential to miRNA

I am rereading this paper and wonder out loud…
http://www.pnas.org/content/107/45/19137.full.pdf

What is the latest we know regarding the status of the common ancestor to gnathostomes & cyclostomes?

Is there credible evidence that modern so-called jawless fish in fact had a jawed and perhaps even boney fish ancestor?

Tom Mueller said...

... I mean I understand the point about "degeneration". I just want to know if this "degeneration" also included the loss of jaws and what evidence would support that contention.

I do hope I am not betraying hopeless confusion again.

Tom Mueller said...

High school students need to learn that there's a lot more to life than mushrooms, maple trees, and moose.

LOL!

Bravo!!!

If you want to jump into that morass then you'd be better off discussing the Three Domain Hypothesis and whether archaebacteria really represent a distinct domain of life that gave rise to eukayotes.

As a matter of fact - that is exactly what I do. My students actually can briefly justify that distinction in biochemical terms. Ex phospholipid ether bonds vs. ester bonds etc making Archaea appear almost alien. Of course eukaryotic ribosomes in organelles vs cytoplasm and the intron/exon story common to Archaea and Eukaryotes etc etc justifying the endosymbiotic story is also emphasized.

Or, for that matter, whether humans are a form of fish.

I often awarded bonus marks for proof demonstrating students have independently read Neil Shubin's book.

Nowadays, I just show the video in class.

I am delighted to confirm you and I are on the same page





Tom Mueller said...

Hi again Larry

I was surprised by your answer and needed to think it over some more

If you want to refer to single-cell eukaryotes as "protists" then that's okay with me.

The important point is that students learn of their existence and their diversity. Quibbling about cladistics and the exact relationship of some multicellular algae is distracting and counter-productive for high school students.


The problem with over-simplification is that students walk away with the erroneous notion there exists a Scala Naturae and that modern single cell Protists are primitive ancestors to more evolved animals plants and fungi.

Of course you would never allow such error in your class, but I assure you: many teachers, when left to their own devices with current textbook resources allow exactly those kinds of ignorachio elenchi to occur.

John Harshman said...

I don't see that this paper, or the one it primarily references, says anything much about the common ancestor. Nor do the fossil data. The record is highly biased toward bones. Euconodonts have something resembling bone and perhaps calcified cartilage. "Agnathans" have dermal bone and sometimes perichondral bone, but never teeth or endochondral bone. Hard to make much of that. But there is no fossil record of bony jaws before gnathostomes.

Tom Mueller said...

Hi John

You were of great help to me earlier...
http://sandwalk.blogspot.ca/2014/04/asbmb-core-concepts-in-biochemistry-and.html

I was wondering about more reinterpretation along these lines extended to include cyclostoma

http://www.scientificamerican.com/article/modern-sharks-may-not-be-living-fossils-after-all/?&WT.mc_id=SA_WR_20140423

John Harshman said...

What do you want? A recognition that both cyclostomes and gnathostomes have evolved since their common ancestor? That should be obvious.

Tom Mueller said...

I was hoping for something along the lines that the last common ancestor to both cyclostomes and gnathostomes had to be a boney fish with jaws ... That would be very nice.

;-)

I was wondering out loud how far "degeneration" could be pushed in this instance

Tom Mueller said...

... But rereading the above, I realize that you already answered my query in the negative.

My apologies

Again thank you for your patience

John Harshman said...

On reconsideration, i do think the fossil record may have something to say here if judiciously combined with time-calibrated molecular trees. If jaws necessarily go along with teeth, then the absence of teeth in the first part of the Paleozoic fossil record tells us there were no jaws either, and my impression is that the cyclostome lineage goes back to before that point. Then again, you can have jaws without teeth, so this must be a tentative conclusion.

Bones, on the other hand, have a somewhat longer record; to my knowledge the oldest bone is Anatolepis, a tiny fragment of dermal bone from the Late Cambrian. Clearly this record is as fragmentary as the fossil itself, but it does suggest that bone precedes teeth in evolution. And, probably, jaws. And, also probably, that cyclostomes are primitively boneless, if they're as old as trees suggest.

Tom Mueller said...

Hi John and again thanks!

This is probably going to be of my most naïve questions ever.

It would appear Cyclostomes loss of miRNA would indicate some sort of “degeneracy” from an ancestral form given the important role miRNA plays in development.

Is it even remotely conceivable to reconstitute the presumed ancient cyclostome miRNA repertoire by genetic engineering in vivo - and see what happens?

Maybe considerations of expression timing and gene dosage are too crucial, making this line of conjecture more the realm of science fiction than reality.

John Harshman said...

First off, every taxon loses miRNAs. That's one of the ways that development evolves. It has nothing in particular to do with "degeneracy". Note that in the study you're talking about, none of the sampled families were lost in cyclostomes, while one was lost in the ancestor of gnathostomes. Now, this is a biased sample, but it does point out that gains and losses are sprinkled all over.

Second, I doubt that reconstructing just the ancestral miRNAs, even if possible, would produce a viable organism, as all sorts of genes and gene products interact in development. Who knows what else you'd have to reconstruct?