In the comments of my dinosaur genome size post, Shelley asked:
So do ALL birds have equally small genomes or is there variation among species?
I don't think she was looking for a trite response along the lines of: "Of course there's variation among species." What she was asking, I presume, is how much variation in genome size do we see in birds? As you can see in this phylogeny, all birds (and nearly all theropods) have small genomes. But that tree only presents data from a few species.
To get a better idea of genome size variation within birds, I downloaded C-values (amount of DNA in a haploid cell measured in picograms) from the Animal Genome Size Database (the same database used by the authors of the nature paper). I included all available data from aves (birds), crocodylia (crocodiles and alligators), tuatara (lizard-like reptiles), squamata (lizards and snakes), and testudines (turtles, tortoises, and terrapins). In some cases, multiple estimates were available for a single species; I averaged those values so that I had a single estimate of genome size for each species. There was only a single species sampled from tuatara (there are only two extant species in this taxon) and only five crocodilians, so these taxa were excluded from much of the analysis.
The results of the analysis are shown below.
In the histogram shown below, you can see that birds (N=204, mean C-value=1.45, SD=0.16) tend to have smaller genomes than squamates (N=253, mean C-value=2.11, SD=0.53) and testudines (N=44, mean C-value=2.86, SD=0.73). The variation in testudines is quite large, whereas birds have very little variation.
For those of you interested in such things, an ANOVA for the effect of taxon on genome size yields a very large test statistic (F=235.53, p
So, Shelley, there ain't much variation in genome size among birds. Purifying selection, anyone?
- Log in to post comments
I think i messed up the link to the boxplot, it lives here:
http://img.photobucket.com/albums/v387/science_boy/Batgenomes.jpg
Something that keeps bugging me about this concept (that birds have smaller genomes and that is selected for because it keeps physical weight down). Birds (and reptiles) have nucleated red blood cells, whereas mammals have anucleate RBC. A whole nucleus has gotta be heavier than a few chunks of DNA within the genome inside that nucleus, yet no one is proposing the equally-logical idea that mammals eliminated RBC nuclei to save weight; nor are they arguing (that I've seen) that bird RBC have nuclei for some flight-related, or even functional, reason.
The weight-loss concept of genome size just kind of strikes me as a desperate attemtp to assign a function to a random ancestral variation. I'd like to see some kind of experimental evidence for it.
Now to the next question: why do avian genomes, though small, reside on gazillions of tiny chromosomes instead of a few big ones?
A couple relevant items I didn't have room for in my Science article...
1. Flightless birds have bigger genomes than flying birds. This is consistent with the idea that metabolism is linked to genome size.
2. In response to iayork's point about mammal red blood cells with no nuclei, it should be pointed out that this issue has been addressed in the arguments for a metabolism genome link. (Go to genomesize.com to find some review paper pdf's.) In fact, these scientists have indeed argued that losing the nucleus may have been an alternative strategy for shrinking red blood cells for warm-blooded metabolism. And here's a strange fact--from mammal to mammal,the size of red blood cells correlates with genome size--even though they have no nucleus.
The authors of the paper actualy address iayork's point -- it's not the weight of the genome, but the metabolic costs of larger genomes. I quoted the relevent passage in my previous post:
Are the same effects seen in flying mammals? I remember an article about "flying DNA" in bats I have seen in the 90s. Unfortunately, I can't find it in the moment.
Don't make me dig up the data from mammals....
RPM,
I'll save you the trouble and make a quick and dirty boxplot .
Looks like bat genomes are litte but not crazily so. Of course that presumes that Chiroptera is a good group (don't know what the prevailing feeling is there at the moment)- maybe if you pulled out echolocating bats (or fuitbats) from the rest you'd see a more pronounced effect.