What is meant by 'random' in DNA mutation?

[Graeme M]

In a recent thread I posted there was mention of evolution having a specific sense in biology. I just spent an hour reading some evolution primers and I think I have a better sense of how this works at the molecular level but will need to read a lot more till it all hangs together in my mind.

However, one thing that immediately stood out is a question about randomness of mutations. I did a little research but couldn't quickly find an answer - most discussions about randomness seemed to focus on rates and locations of change. I think I have misunderstood what is meant by random.

My question is pretty basic. If mutations arise in a random fashion and then can be selected for or against, how is it that the same deletrious mutations may be found on an ongoing basis? I apologise if my terminology is not correct, this is from just the briefest of introductions to the topic.

What I mean is this. If a mutation to a gene gives rise to an abnormality that decreases an organisms fitness, how is it that the mutation is preserved over time? I am thinking of for example a genetic disease. If a creature has a genetic condition that reduces its fitness wouldn't it follow that the particular mutation would be selected against and eventually disappear from the gene pool? If mutation is random, and selective pressure acts to preserve advantageous changes but not disavantageous changes, shouldn't the case be that a particular condition does not remain evident over time? Or that we should be seeing variations of a condition over time?

 

[Evo]

In a recent thread I posted there was mention of evolution having a specific sense in biology. I just spent an hour reading some evolution primers and I think I have a better sense of how this works at the molecular level but will need to read a lot more till it all hangs together in my mind.

However, one thing that immediately stood out is a question about randomness of mutations. I did a little research but couldn't quickly find an answer - most discussions about randomness seemed to focus on rates and locations of change. I think I have misunderstood what is meant by random.

My question is pretty basic. If mutations arise in a random fashion and then can be selected for or against, how is it that the same deletrious mutations may be found on an ongoing basis? I apologise if my terminology is not correct, this is from just the briefest of introductions to the topic.

What I mean is this. If a mutation to a gene gives rise to an abnormality that decreases an organisms fitness, how is it that the mutation is preserved over time? I am thinking of for example a genetic disease. If a creature has a genetic condition that reduces its fitness wouldn't it follow that the particular mutation would be selected against and eventually disappear from the gene pool? If mutation is random, and selective pressure acts to preserve advantageous changes but not disavantageous changes, shouldn't the case be that a particular condition does not remain evident over time? Or that we should be seeing variations of a condition over time?

No, just because a mutation is not advantageous does not mean that it is selected against and stopped. If the person with the mutation breeds and their children breed, the mutation will likely be passed on in some generation. Evolution does not just select for overall positive changes. Take sickle cell anemia, for example.

 

[Ygggdrasil]

If a mutation to a gene gives rise to an abnormality that decreases an organisms fitness, how is it that the mutation is preserved over time? I am thinking of for example a genetic disease. If a creature has a genetic condition that reduces its fitness wouldn't it follow that the particular mutation would be selected against and eventually disappear from the gene pool? If mutation is random, and selective pressure acts to preserve advantageous changes but not disavantageous changes, shouldn't the case be that a particular condition does not remain evident over time? Or that we should be seeing variations of a condition over time?

This is a good question that many studying evolution often bring up.

Most genetic diseases are due to inherited mutations that arose many generations in the past. These include many fairly well-known genetic disorders like Huntington's disease, sickle cell anemia, and cystic fibrosis. Why does evolution not weed out these mutations from the gene pool? In some cases, the selection against the disease is not strong enough to weed out the disease. This is the case for Huntington's disease, whose symptoms usually begin after the typical age of reproduction. In other cases, the mutation associated with the genetic disease has beneficial side effects. For example, while the sickle-cell allele causes sickle cell disease when present in two copies, but it is associated with protection against malaria in individuals that carry only one copy of the allele.

There are some genetic disorders that are caused by de novo mutations, mutations that were not present in one's parents, but these are fairly rare. Sometimes conditions can be cause by a number of different mutations in related genes. Many cancers are associated with de novo mutations, but there are often a number of different mutations in related proteins that are associated with the disease rather than a small number of mutations in one gene as seen in most inherited genetic disorders.

 

[Graeme M]

Thank you for that clarification, that is very interesting that a deleterious mutation might in some circumstances also confer an advantage. In terms of my question though I mean something slightly different.

I am not talking necessarily about human beings, rather the mechanism in general as applied to any population of organisms. If a mutation results in a disorder that significantly impairs the organism I assume there is a chance the creature will not reproduce. For example a disorder that results in substantial motor impairment.

If that mutation causes disorder A which prevents the organism reproducing, then has that mutation been actively selected against and removed from the gene pool? And if so, what is the likelihood that exact mutation will occur again in the population? In other words, are some, or many, particular deleterious mutations likely to occur in a population even though not inherited?

Such a mutation would not be extant in any population, yet would have a likelihood of occurring n times in any sample.

 

[Ryan_m_b]

If that mutation causes disorder A which prevents the organism reproducing, then has that mutation been actively selected against and removed from the gene pool?

If a mutation lowers fitness, that is prevents or lowers the rate of reproduction, then by definition it is deleterious. However not all deleterious mutations are the same. Some are strongly deleterious which means they are removed from the gene pool quickly (e.g. a mutation that makes one infertile) but others linger because the effect on reproduction is minor. In addition some mutations that seem deleterious, like peacock tails that make moving difficult and attract parasites, are actually advantageous because they confer a greater advantage.

And if so, what is the likelihood that exact mutation will occur again in the population? In other words, are some, or many, particular deleterious mutations likely to occur in a population even though not inherited?

Such a mutation would not be extant in any population, yet would have a likelihood of occurring n times in any sample.

I'm not sure how likely that is however there are areas of the genome that are more prone to mutation to others due to epigenetic factors. This paper covers why some sections of DNA in sperm are more prone to errors, have a read and see what you take from it.

 

[DiracPool]

If that mutation causes disorder A which prevents the organism reproducing, then has that mutation been actively selected against and removed from the gene pool?

I think that goes without saying, doesn't it? Is pretty obvious.

And if so, what is the likelihood that exact mutation will occur again in the population?

You're speaking in very general terms. The title of your thread refers to "random" mutations. The likelihood that that exact mutation will occur again is, well...random, in that case.

In other words, are some, or many, particular deleterious mutations likely to occur in a population even though not inherited?

I'm not sure what you mean by this. Are you talking about a situation whereby a deleterious mutation occurs in a population and is negatively selected out of that population and then reappears every so often with a certain frequency because the genome is particularly susceptible to this certain type of mutation? Try to be a bit more specific.

 

[Drakkith]

In addition some mutations that seem deleterious, like peacock tails that make moving difficult and attract parasites, are actually advantageous because they confer a greater advantage.

Indeed. I expect it is quite difficult to determine whether some mutations are wholly deleterious or if they actually confer some other benefit that is just very difficult to see.

 

[Graeme M]

Thanks Ryan, I'll have a read.

I'm not sure what you mean by this. Are you talking about a situation whereby a deleterious mutation occurs in a population and is negatively selected out of that population and then reappears every so often with a certain frequency because the genome is particularly susceptible to this certain type of mutation?

Yes, that's largely what I meant. I had assumed that mutations occur with no particular frequency or likelihood, in fact I'd have imagined that a particular mutation might never occur more than once.

So my question just means - if an organism is born with a mutation that prevents it reproducing, that mutation will not remain in the gene pool. How likely is it that the exact mutation will occur again? Or put at the gross level, if an organism is born with disability A that prevents it reproducing before it dies, what are the chances that disability A will reappear in the population.

 

[DiracPool]

So my question just means - if an organism is born with a mutation that prevents it reproducing, that mutation will not remain in the gene pool. How likely is it that the exact mutation will occur again? Or put at the gross level, if an organism is born with disability A that prevents it reproducing before it dies, what are the chances that disability A will reappear in the population.

I'm basically in Ryan_m_b's camp on this one...

I'm not sure how likely that is however there are areas of the genome that are more prone to mutation to others due to epigenetic factors.

I never really gave it much thought or even know to what extent research is being done on that issue. Is there some practical reason you are interested in this question, or is it just a general curiosity?

 

[Ygggdrasil]

If that mutation causes disorder A which prevents the organism reproducing, then has that mutation been actively selected against and removed from the gene pool?

A few notes on this point. Most genetic diseases are caused by recessive alleles, so an individual needs to have two copies of the allele in order to show signs of the disease. People with one copy of the allele are perfectly healthy or, in cases like sickle-cell anemia, may even have advantages. Nevertheless, when a recessive allele becomes very rare in a population (as most disease causing alleles are), there are very few individuals who carry two copies of the allele and show signs of disease. The vast majority of people who carry the allele have only one copy and have no survival disadvantage. Thus, as recessive, disease-causing alleles become rare in a population, the selection against the allele gets very weak, so natural selection is not very good at completely eliminating these alleles from the population.

And if so, what is the likelihood that exact mutation will occur again in the population? In other words, are some, or many, particular deleterious mutations likely to occur in a population even though not inherited?

As I mentioned in my previous post, most diseases associated with de novo mutations are not caused by the exact same mutations that recur at the same location. Rather, every individual with the disorder likely carries a different mutation. Autism is one example of a disorder thought to come about partly from de novo mutations. When researchers looked through sequencing data from a study of ~500 individuals with autism, they only found 18 genes that were mutated in multiple patients (see the section on recurrently mutated genes in http://www.nature.com/nrg/journal/v13/n8/full/nrg3241.html), and these recurrently mutated genes probably had different mutations in different patients.

That said, there is some suggestion that certain regions of the genome may have higher mutation rates than others (forming "mutation hot spots"). There are some thoughts as to how this may come about, but the processes (and maybe even the existence of mutation hotspots) is still being debated and studied. There are certainly cases of mutations that frequently arise de novo in cancer, however. One example is the "Philadelphia" chromosome rearrangement that fuses parts of chromosomes 9 and 22 to cause most cases of chronic myeloid leukemia (see http://www.molecular-cancer.com/content/9/1/120).

Here's some further reading on mutation hot spots in case anyone is interested:
Drake JW 2007 Too Many Mutants with Multiple Mutations. Crit Rev Biochem Mol Biol 42: 247. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2265383/

Shee, Gibson and Rosenberg 2012. Two Mechanisms Produce Mutation Hotspots at DNA Breaks in Escherichia coli. Cell Rep. 2:714. http://dx.doi.org/10.1016/j.celrep.2012.08.033[/URL]

 

[Graeme M]

Yes, the idea of 'hotspots' is pretty much what I was getting at.

DiracPool, no practical application, just curiosity. It's what immediately struck me when I was reading about the basic mechanics. Remember, I have no more than a passing acquaintance with the whole idea from basic high school science and some reading of popular books (and now an hour or so of internet research) so I am coming from a position of considerable ignorance (as might have been implied earlier - I wouldn't disagree!).

Thanks for the various suggested references, I hope to get to and read these soon. The first one from Ryan has some great related links and at least one of those has some interesting comments. I gather from a quick and superficial reading that the idea of hotspots, while not exactly controversial is nonetheless not regarded as mainstream as yet, would that be correct?

Does anyone know whether there is research/ideas about why there should be such things as hotspots? I'll chase down the various references and see what I find, anyway.

 

[Ygggdrasil]

Thanks for the various suggested references, I hope to get to and read these soon. The first one from Ryan has some great related links and at least one of those has some interesting comments. I gather from a quick and superficial reading that the idea of hotspots, while not exactly controversial is nonetheless not regarded as mainstream as yet, would that be correct?

I'm not so familiar with this area of research, so I don't know the how much is accepted and how much is still being worked out. A quick check of Molecular Biology of the Cell (5th edition, 2008), one of the standard textbooks for molecular and cell biology, doesn't include any mention of the topic, which indicates that it is fairly new area of research and our understanding of the topic is still in flux.

Does anyone know whether there is research/ideas about why there should be such things as hotspots? I'll chase down the various references and see what I find, anyway.

The Cell Rep. paper that I linked in my previous post discusses some potential explanations for why hotspots exist in bacteria. Here's a paper suggestion how the proteins that package DNA in eukaryotes influences mutation rates:

Chen et al. 2012. Nucleosomes Suppress Spontaneous Mutations Base-Specifically in Eukaryotes. Science 335: 1235. doi: 10.1126/science.1217580

 

[Laroxe]

I think the copying of DNA is an imperfect process and I've read that the average person has as many as 300 differences in the DNA from their parents. Most of these copy errors appear not to have much of an effect, the ones that do are more likely to be harmful than useful and for any to be positively selected its highly dependent on the environment and a lot of luck. Epigenetics is really about gene expression rather than any permanent changes in DNA so is unlikely to be a big player. However there are a few new ideas that suggest some external influences on our genome, one is "Nutrient-dependent/pheromone-controlled adaptive evolution", which to be honest, I have yet to get my head around. The other is linked to the finding that human DNA appears to have inclusions from viral and bacterial sources that have become integrated and are transmitted to our offspring. Both of the mechanisms suggest that evolutionary change may occur over a much shorter timescale than previously thought and might also reflect the fitness needs of none human organisms. These ideas do suggest that the processes may not be quite as random as previously thought.

 

[DaveC426913]

Indeed. I expect it is quite difficult to determine whether some mutations are wholly deleterious or if they actually confer some other benefit that is just very difficult to see.

All the more so because it is subjective - and emergent.

For many mutations, there is no test - even in principle - by which one can, a priori , say 'A will be beneficial' or 'B will be harmful'. That is only for time - and population counts - to tell.

It sucks to be anemic, but no one could have predicted that it would actually confer a survival advantage via lower risk for malaria.

 

[write4u]

All the more so because it is subjective - and emergent.

For many mutations, there is no test - even in principle - by which one can, a priori , say 'A will be beneficial' or 'B will be harmful'. That is only for time - and population counts - to tell.

It sucks to be anemic, but no one could have predicted that it would actually confer a survival advantage via lower risk for malaria.

A perfect example of a beneficial mutation may be found in the human brain.

All great apes apart from man have 24 pairs of chromosomes. There is therefore a hypothesis that the common ancestor of all great apes had 24 pairs of chromosomes and that the fusion of two of the ancestor's chromosomes created chromosome 2 in humans. The evidence for this hypothesis is very strong

http://www.evolutionpages.com/chromosome_2.htm

 

[Graeme M]

That is quite remarkable. Even more so is the suggestion that viable offspring can be produced from inter-species breeding which is something I always thought didn't happen.

I don't know why you say it may be found 'in the human brain'? Is chromosome 2 linked in some way to brain development in the human line?

 

[write4u]

That is quite remarkable. Even more so is the suggestion that viable offspring can be produced from inter-species breeding which is something I always thought didn't happen.

I am not suggesting that apes can interbreed with humans, just that we had a common ancestor who had 24 chromosomes and the mutation produced a hybrid, early man, who had 23 chromosomes from the fusion of two chromosomes in our common ancestor.
I am not expert enought to know if this single genetic difference excludes interbreeding. Some other species can, i.e. the mule.

A mule is the offspring of a male donkey (jack) and a female horse (mare). Horses and donkeys are different species, with different numbers of chromosomes.

.

I don't know why you say it may be found 'in the human brain'? Is chromosome 2 linked in some way to brain development in the human line?

We are almost identical in genetics, but what sets humans apart from other Hominids is our brain. Thus, if the mutation produced man, it would certainly include a bigger brain and reasoning powers. Any other difference in appearance from our ancestor in both species is a result of natural environmental evolutionary selection, not a gene altering mutation.
In addition, there is no evidence that early man was less intelligent than modern man. They just lacked knowledge. This would seem to indicate that there is a good chance that this genetic mutation affected the growth instructions of the brain itself. Perhaps a folding instruction which allowed for a more complex neural network

 

[write4u]

Thanks Ryan, I'll have a read.

Yes, that's largely what I meant. I had assumed that mutations occur with no particular frequency or likelihood, in fact I'd have imagined that a particular mutation might never occur more than once.

So my question just means - if an organism is born with a mutation that prevents it reproducing, that mutation will not remain in the gene pool. How likely is it that the exact mutation will occur again? Or put at the gross level, if an organism is born with disability A that prevents it reproducing before it dies, what are the chances that disability A will reappear in the population.

It occurred to me that you may be looking at this from the wrong viewpoint. Once the mutation has occurred it IS part of the gene of the offspring and will be passed on in their offspring with regularity.
If the mutation is somehow prevented from being passed on, then that mutation dies. But in view of the enormous complexity of cell division it is unlikely that any mutatations occur with any regularity, let alone the same mutation.

An excellent representation of gene duplication: http://www.ted.com/talks/drew_berry_animations_of_unseeable_biology

 

[Graeme M]

As always I apologise if my level of understanding means I ask silly questions. I have found the discussion illuminating and the references quite interesting and educational. But my actual knowledge of the mechanics is somewhere between none and zip...

However, I assumed from the link that the mutation to produce Chromosome 2 would have arisen once only (is this a de novo mutation). The resulting being then had to reproduce. Whatever advantage this chromosomal fusion offered was of no value if the being did not reproduce. I sort of had imagined that differing numbers of chromosomes would prevent that being from producing viable offspring if he/she mated with another ape.

And that goes to the heart of my earlier question. Again, I think my attempts to describe what I mean are missing the mark - most likely because I have over-simplified the process in my mind.

You note that the mutation IS part of the offsprings genes and yes, I understood that. You then say that the mutation 'dies' if the offspring does not produce its own offspring. My question was simply that if a particular mutation gives rise to a particular physical trait (is that phenotype?) and that trait prevents successful reproduction, then the trait is lost from the population (it is not heritable). How likely is it though that the particular mutation will arise again by chance?

Or in practical terms using the example you give, what are the chances that a fusion of chromosome 2p and 2q might have occurred again over the many thousands of years since it obviously did happen? Or from another angle, could that particular mutation occur multiple times in the set of apes that have ever lived?

 

[write4u]

As always I apologise if my level of understanding means I ask silly questions. I have found the discussion illuminating and the references quite interesting and educational. But my actual knowledge of the mechanics is somewhere between none and zip...

However, I assumed from the link that the mutation to produce Chromosome 2 would have arisen once only (is this a de novo mutation). The resulting being then had to reproduce. Whatever advantage this chromosomal fusion offered was of no value if the being did not reproduce. I sort of had imagined that differing numbers of chromosomes would prevent that being from producing viable offspring if he/she mated with another ape.

And that goes to the heart of my earlier question. Again, I think my attempts to describe what I mean are missing the mark - most likely because I have over-simplified the process in my mind.

You note that the mutation IS part of the offsprings genes and yes, I understood that. You then say that the mutation 'dies' if the offspring does not produce its own offspring. My question was simply that if a particular mutation gives rise to a particular physical trait (is that phenotype?) and that trait prevents successful reproduction, then the trait is lost from the population (it is not heritable). How likely is it though that the particular mutation will arise again by chance?

Or in practical terms using the example you give, what are the chances that a fusion of chromosome 2p and 2q might have occurred again over the many thousands of years since it obviously did happen? Or from another angle, could that particular mutation occur multiple times in the set of apes that have ever lived?

Could such a mutation occur today in, say a Bonobo Chimp, and produce a more peaceful intelligent hominid, I suppose it is possible, but as I understand it any significant mutation is extremely rare (see above link to the process).

 

[Torbjorn_L]

However, one thing that immediately stood out is a question about randomness of mutations. I did a little research but couldn't quickly find an answer - most discussions about randomness seemed to focus on rates and locations of change. I think I have misunderstood what is meant by random.

My question is pretty basic. If mutations arise in a random fashion and then can be selected for or against, how is it that the same deletrious mutations may be found on an ongoing basis? I apologise if my terminology is not correct, this is from just the briefest of introductions to the topic.

I remember having a hard time with this too when I started to get interested in evolution, later astrobiology. How does the different distributions (say, recurrent mutations) add up to "random" and what does it mean?

However I found out that "random" is a lazy description here. What is necessary for evolution is that there is a genetic variation (i.e. also sexual recombination count) among a population for selection to work on and that this is a causal process. E.g. variation can't "know" what selection will make of it - unless some mechanism can teleport information and back in time to boot. Some biologists [Coyne, for one] describe variation as "indifferent" to selection, which seem to be as apt as a population genetic mechanism can be described in words.

The take home message for me was that variation could have any odd distribution as long as it is based on a stochastic (causal) process. 'Dr. Strangekitty or: How I Learned to Stop Worrying and Love the Mutational Bomb'.

 

[write4u]

I remember having a hard time with this too when I started to get interested in evolution, later astrobiology. How does the different distributions (say, recurrent mutations) add up to "random" and what does it mean?

However I found out that "random" is a lazy description here. What is necessary for evolution is that there is a genetic variation (i.e. also sexual recombination count) among a population for selection to work on and that this is a causal process. E.g. variation can't "know" what selection will make of it - unless some mechanism can teleport information and back in time to boot. Some biologists [Coyne, for one] describe variation as "indifferent" to selection, which seem to be as apt as a population genetic mechanism can be described in words.

The take home message for me was that variation could have any odd distribution as long as it is based on a stochastic (causal) process. 'Dr. Strangekitty or: How I Learned to Stop Worrying and Love the Mutational Bomb'.

Thank you for clarifying what I was attempting to posit.

I was considering a "mutation" to be a catastrophic event to the duplication function, such as an injury. This needs not necessarily be detrimental, but it most often is. It seems to me that, for instance, a fusion of two genomes into a single genome twice its size would normally be a catastrophic event leading to significant aberrations. But in the example above it is clearly the event from which man emerged and the significant aberration was ultimately beneficial. That's why it survived the test of natural selection, at least until now.

 

[Torbjorn_L]

Thank you for clarifying what I was attempting to posit.

I was considering a "mutation" to be a catastrophic event to the duplication function, such as an injury. This needs not necessarily be detrimental, but it most often is.

Secretly mutations, a variation of DNA bases, are trivially neutral in ~ 30 % of cases, since the translation code is that much redundant. So it ain't all bad.

In fact the variation of sexual recombination is a basis for helping eliminate the effects of other change (when selection is applied).

 

[write4u]

Secretly mutations, a variation of DNA bases, are trivially neutral in ~ 30 % of cases, since the translation code is that much redundant. So it ain't all bad.

In fact the variation of sexual recombination is a basis for helping eliminate the effects of other change (when selection is applied).

I understand the redundancy being necessary for correction of random imperfections. But what if the mutation can no longer be corrected but replicates as per instruction of the mutated DNA.

All great apes apart from man have 24 pairs of chromosomes. There is therefore a hypothesis that the common ancestor of all great apes had 24 pairs of chromosomes and that the fusion of two of the ancestor's chromosomes created chromosome 2 in humans. The evidence for this hypothesis is very strong

http://www.evolutionpages.com/chromosome_2.htm

 

[Pythagorean]

I think random here simply means that there's an element of change that changes independently of the current state of the system, where state is not the exact physical geometry, but only the organization of the genetic code as we interpret it. The mutations, realistically, are dependent on the physical state of the system, but at a scale we can't appreciate very well.

 

[rootone]

Whatever the cause of a mutation may be, what it means physically is that the sequences of bases in some part of the DNA gets altered.
There are a number of known reasons why this can happen, viruses being one example, and EM radiation being another.
Most of the time such a mutation results in a non functional or possibly harmful protein being produced instead of what was originally encoded.
Occasionally though, the new encoding 'luckily' produced an improvement instead.

 

[Torbjorn_L]

I understand the redundancy being necessary for correction of random imperfections. But what if the mutation can no longer be corrected but replicates as per instruction of the mutated DNA.

Neutral mutations are not visible to selection, so they are not "corrected" but propagated.* Hence it is possible that a following base pair change may be more severe than if the first didn't happen. (Change to a different amino acid.)

[*Something similar happens for so called "near neutral mutations" in real life populations. Population genetics can show that you need ever larger populations and time to pick up variations with ever smaller effects and either promote or eliminate them. But I mention this to be complete, it isn't necessary to know about it for understanding variation as such.]

I think "correction" is the wrong term here. First because variation is inherently neither bad nor good, it is simply necessary for evolution, which is what makes populations survive a range of sufficiently large environmental change to lead to extinction else.** Second because there are DNA repair mechanisms for broken strands and what not, so I would use "repair" in that case.

[** And to be complete again, some changes are too large or too rapid to be mooted that way. I am just learning more about the Great Dying, the end-Permian extinction, where 90 % of all organisms disappeared. I wasn't troubled before, because life has survived so many mass extinctions. (And the previous Great Oxygenation Event was probably an even larger killer, but we can't see that yet.)

But it took unusually long before the wound healed, 10+ million years for ecological balance and _100+_ million years for recovery of diversity. (When most mass extinctions are recovered on both counts after 1-10 million years or so.) And the species that survived seemed to have been precisely random outcomes! No discernible patterns of "better" robustness. (Compare with the KPg extinction that hit herbivorous and carnivorous dinosaurs, while insectivorous mammals and perhaps insectivorous avian dinosaurs survived, so a distinct pattern.)

It appears a diversified biosphere helps but is not the ultimate answer to all change. And of course when our star eventually starts to die, so does life... But now I appreciate the limits of evolution more.]