So… no @RonSewell, Bill can’t summarize what people have repeatedly told him.
6 Likes
-
Could you please state in your own words what this “waiting time” problem is and how it would prevent the pattern of gains and losses we see?
-
Hasn’t that, if true, made gene loss easy? And, in a gene that’s needed by the species, wouldn’t selection prevent such degradation?
-
What do you mean by “large gene changes”? The magnitude of changes is nowhere shown in the diagram. If you mean lots of gains and losses, that has nothing to do with large changes to the sequences.
He showed no such thing, and the Howe diagram fits the nested hierarchy almost perfectly. Is that expected from separate creation?
Not true. Only a very few don’t follow the hierarchical pattern. This suggests that you have no understanding of the figure.
3 Likes
It’s the time for mutations to become fixed in a population. Models that include only a few mutations take millions of years. There is over 10k non shared genes between the 4 species in the Howe.
More gibberish. I can tell you don’t understand any of this. I’ll try to explain with a rough model, and I’ll assume neutral evolution for simplicity. If there’s selection, everything takes less time. First, the average time from mutation to fixation for an allele that eventually becomes fixed is 4N generations. That’s a long time. But then again, the number of mutations fixed in every generation is equal to the mutation rate per individual per generation. The mutations that get fixed in this generation, say around 100 in the human species, all happened long ago, but that many are still being fixed every generation. Most mutations, of course, never become fixed and never reach a high frequency, but the percentage fixed and the time it takes to fix them are irrelevant to the number fixed. Now, I have no good estimate for the rate of mutations that eliminate or create a gene, but we can consider what the mutation rate would have to be. If there are 10,000 unshared genes, let’s imagine that half of them are gains and half are losses. In either case, that could involve a single mutation each. For either, then, that’s 5000 fixations. Supposing that the split between human and zebrafish is 400ma, between human and chicken is 250ma, and between human and mouse is 70ma. Add up the tree, and that’s 1130ma total lineage time. That’s one mutation of each sort every 226,000 years, give or take, or one in every 226,000 individuals per year. Is that an impossible number? What’s your argument?
7 Likes
Is the claim that you do not have enough data to test the hypothesis so your rough calculations are a feasibility estimate?
How did you arrive at this estimate?
How did you arrive at the number 1130ma?
Probably via a typo - it should be 1120ma.
Test what hypothesis? Your clarity suffers when you don’t know what you’re talking about, as is generally the case.
It’s not an estimate. It’s a fact. a gene can turn into a pseudogene by a single mutation that adds a stop codon or deletes a base. A new gene can arise from a single duplication. These are only examples, but perhaps you might get the idea.
By adding up all the branches of the tree. I see, however, that I used the date for separation of archosaurs and lepidosaurs rather than the date for separation of synapsids and sauropsids. I’ll just look up a bit. Human from zebrafish: 430ma; zebrafish from human: same; human from chicken: 320ma; human from mouse: 90ma. Add them all up: 1370ma. So one every 274,000 years.
2 Likes
That Venn diagram again… Didn’t expect it so soon after the last time. Did someone buy that as an NFT?
3 Likes
So… you don’t even know what the waiting time problem is. Fun.
Of which only 130 are interesting, and none are surprising.
3 Likes
Hi, Bill, I look forward to your response! Thanks!
1 Like
The hypothesis that gene gain and gene loss explain the Howe venn pattern.
Your estimate is built on unrealistic assumptions.
I still dont she how you got 1370 but until your assumptions are realistic this number is trivial.
Hi Matt
I think this is fine.
1 Like
There are hundreds of human olfactory pseudogenes. Do you think that is unrealistic?
Human specific loss of olfactory receptor genes
Olfactory receptor (OR) genes constitute the basis for the sense of smell and are encoded by the largest mammalian gene superfamily of >1,000 genes. In humans, >60% of these are pseudogenes. In contrast, the mouse OR repertoire, although of roughly equal size, contains only approximately 20% pseudogenes.
3 Likes
Nope, that’s not what this is testing. The evidence for that is the nested hierarchy.
What would be more realistic assumptions? How do you know what’s realistic?
This should be a signal that you don’t understand any of this stuff. The branches of a tree represent time. the distance from the tips to the root is 430 million years, the age of the divergence of the human/mouse/chicken clade (Sarcopterygia) from the zebrafish clade (Actinopterygia). You just add up the lengths of the branches to get the total evolutionary time allowed for gene gains and losses. How is that unclear?
2 Likes
That adds up to 1270.
Anyway, @colewd this is it. Mere addition of branch lengths:
Not that complicated.
1 Like
Hi Ron
Thanks for the citation. The presents of pseudogenes is quite interesting. The cause is a much more challenging problem.
From the abstract:
We found that humans have accumulated mutations that disrupt OR coding regions roughly 4-fold faster than any other species sampled. As a consequence, the fraction of OR pseudogenes in humans is almost twice as high as in the non-human primates, suggesting a human-specific process of OR gene disruption, likely due to a reduced chemosensory dependence relative to apes.
This change would not be expected based on neutral theory. Why would loss of smell have such strong selective pressure?
In table 1 there are the symbols stop and deletion. Is this where the loss is due to a premature stop condon vs a deletion of the gene?
You have it backwards. It isn’t loss of smell being selected for in humans, it’s retention of smell being selected for in non-humans.
Yes, if you mean “stop codon”. But the gene doesn’t have to be deleted, just some piece of it, especially if it’s out of frame.
2 Likes
You will note that there is a third symbol, insertion. Most likely also out of frame.
It may be more a case of loss of purifying selection as other factors reduce dependency on particular smells, including shifts in environment and ecology, and increased reliance on other sensory inputs, such as vision.
Genes do not generally just disappear without a trace in one go. Pseudogenes are no longer protected by purifying selection and are subject to further erosion.
3 Likes
Yes it would. It is exactly the lack of a function-maintaining selective pressure that allows deactivating mutations to accumulate in these genes.
There isn’t a selective pressure for the loss of smell, rather there is a lack of selective pressure for a more sensitive sense of smell at least with respect to certain smells.
Nine OR genes were intact in all primate species examined (Table (Table1).1). Along the human evolutionary lineage, only one amino acid change has occurred in the putative OR-binding sites [a total of 261 amino acids (19); O.M., Y.G. and D.L., unpublished results] of these nine OR genes. This compares with 11 changes in the 14 OR genes (406 amino acids) intact in humans but not in one or more of the other primates examined (P = 0.034). This observation may suggest that evolutionary constraints differ among human OR genes. We propose that OR genes in the human genome belong to three functional groups: (i ) OR genes that are essential to all primates and therefore are under selective pressure to remain intact in humans as well; (ii ) OR genes that are not important for humans but are essential for other primates; (iii ) OR pseudogenes that have lost their function in humans. The two latter categories of genes would accumulate coding region disruptions at a neutral rate in humans, thus explaining the high rate of OR gene silencing observed in the human evolutionary lineage.
Yes it’s premature stop codons and deletions at a particular position in the coding sequence. The total amount of deleted sequence isn’t shown so it can be anything from a single nucleotide(resulting in a frameshift mutation), to the remainder of the gene.
4 Likes