quote:
Originally posted by John Paul:
1. Do you believe that mutations are heritable?
2. Do you believe that the patterns of such heritable mutations can be used to infer relatedness?
Hi JP,
There’s a few misconceptions here that I’d like to tackle, if I may.
Phylogenetic inference is a method which we infer relationships of genetic sequences. It is important to make this distinct from the inferred
species tree from such analysis’. To reliably be able to infer a species tree, we may need several gene/data trees, maybe morphological trees as well, in order to get a reliable enough consensus.
Imagine a species tree to be a glass pipe stretching back into the past. Within this pipe are cotton threads representing gene lineages. Now, imagine a gene for normal colour sight that gets a mutation & becomes the colourblind allele. At some point in the past, if you trace the two threads (alleles) back to the the mutation point, they will join. This is called a coalescent event (it’s arse about face, if you ask me, but in the interests of using the same definitions as everyone else.). As such, gene lineages aren’t expected to completely match species lineages. It gets more complicated if the coalescent event occurred before speciation, because you have one pipe getting itself a Y connector (speciation event), with both alleles splitting & heading up different pipes (separate species), for a total of four threads.
BUT, it matters not one iota whether that gene allowed excellent colour vision or not (I’m assuming, purely for the sake of argument that colour blindness is caused by a mutant allele, I could easily use any gene), because you are tracing back two (or more) homologous alleles (from different species) to the coalescent event, which is what [/I]infers the relationship[/I]. The allele frequencies that cause the inheritance patterns is an utterly separate issue.
As I have said above, gene lineages need not match exactly (in real life) the species lineage, but nails down most of the details. Unresolved parts of the first tree may be resolved by a second & third molecular phylogeny using different genes/data sets. In the end, we can reliably infer a species tree using
multiple congruent phylogenies derived from many different & diverse data sets.
Of course, the above requires sequences that are both different enough to be informative, & homologous. This is why a single polymorphism isn’t adequate to infer very much. As a result, not being able to infer relationships from a sickle cell allele, or trisomy 21, isn’t exactly falsifying molecular phylogenetic inference.
We are attempting to infer species trees from sequence data (cotton threads) that that has survived in a direct unbroken lineage all the way up those pipes, being split into two, & threaded up two pipes (Y connectors) potentially many times. There are threads that don’t make it, & are lost, we can only infer from the extant threads that poke out the top (present day).
If common descent is true, then phylogenetic analysis should consistently show (& there are reasons why it doesn’t) that homologous sequences are related, & subsequent studies should be broadly congruent. They are.
Apologies for the analogy, it just seemed easier.
quote:
Originally posted by John Paul:
1/ Is tricky. Yes mutations are heritable. Neutral, harmful and beneficial, mutations can be passed on. However in sexual reproduction they don’t always get passed on. I’m not as tall as my grandfather was, but I am taller than my parents (were). My father was color-blind, I am not, nor are my sisters and brothers, but I have a nephew that can’t see green (not the Special Agent). However his brother’s vision is OK.
You are confusing inheritance patterns with phylogeny.
Assuming you had sequence data of homologous colour-blindness genes between humans & apes, you could infer a phylogeny, provided that there were enough informative differences between extant sequences, or you had a large variable sample, & then similarities would mean recent divergence.
quote:
Originally posted by John Paul:
In humans this is evident- not every organism that is born gets a chance to mate and not every mating couple can conceive. No mating or conception no chance of passing on of the DNA.
Irrelevant. No passing on of DNA, no phylogeny from which to infer anyway!
quote:
Originally posted by John Paul:
Take an organism born with a beneficial mutation that its parents didn’t have, nor do its siblings. Not only does this organism have to live long enough to reproduce, it has to do so successfully in order just to have a chance of that beneficial mutation being passed on, never mind becoming fixed. Another factor would be having a genetically impaired mate such that any combination would give you offspring less functional than the better parent is (was). You know, basic Punnett Square stuff and Mendelian genetics.
That said, if adaptive mutations were the norm (Dr. Lee Spetner), they would become more readably fixed because they would occur population wide due to the organisms’ DNA reacting directly to environmental pressure(s). However adaptive mutations, unless applied to cleverly written evolutionary algorithm acting with an incrementally sequenced genetic algorithm, couldn’t account for the grand sweep of the theory of evolution.
What we would have to determine is what was it about the alleged shared mutations that allowed them to be fixed in the populations? IOW why were they selected for (kept in the population) over this alleged span of time (5+ millions years)?
Irrelevant. This has nowt to do with inferring phylogenies, except the first sentence that describes a coalescent event.
quote:
Originally posted by John Paul:
2/ I don’t think that every person with sickle-cell anemia is related to the first person that got the mutation that caused that disease and was able to pass it on. (Sickle-cell anemia is caused by a point mutation in a specific locus- a substitution of a T for an A in the codon for the sixth amino acid of the beta chain in the human hemoglobin protein. That mutation changes a glutamic acid to a valine.) Is everyone with Downs syndrome related? The same goes for all genetic diseases. Do you think that every person with the same genetic disease is related to each other? That DNA gets passed on to the offspring doesn’t mean chimps and humans share a common ancestor.
As I previously stated As for apparent similar mutations, again given that we have a restricted selection of possibilities for change to occur, it could be more of a coincidence than it is coinciding. I would like to change that to most likely be more of a coincidence
Regarding sickle cell, you may be right. But since it’s a
single polymorphism that proves to have a net beneficial effect to the owners in malaria infested Africa, it’s entirely possible that the identical alleles arose more than once, incidences of trisomy 21 certainly do. In any case, a single polymorphism across the entire potential range of samples isn’t enough to infer a phylogeny with, except a close relationship with the original heamaglobin.
But that’s not what were talking about, is it? Were talking about homologous sequences between species, not people, that have A LOT in common. Suggesting that these alleles arose spontaneously alike, & were then fixed, neutral loci as well, is an entirely different prospect to a single polymorphism. Not only did they arise spontaneously alike, phylogenetic inference shows that they are congruent with other genetic phylogenies. By chance? The odds are colossally against such an event. An 11 taxa phylogeny has 34,000,000 possible trees. I’ll give you the benefit of the doubt & assume that two phylogenies for the same 11 taxa was
only 50% congruent. That’s 17,000,000 * 17,000,000 = 289,000,000,000,000 : 1 odds that such an event occurred by chance. Because of this, & the fact it occurs over & over, most people reasonably accept that phylogenetic analysis is good evidence of common descent.
quote:
Originally posted by John Paul:
Mutations occurring and getting passed on is just part of the problem. And a mutation getting fixed in a population is another.
Nope. If a mutation doesn’t get passed on, we can’t infer from it. If a mutation isn’t fixed, & worse, is lost, the same applies.
quote:
Originally posted by John Paul:
What the theory of evolution requires is for mutations to accumulate in such a way as to eventually give rise to new structures and organs (assuming of course the alleged starting population(s) didn’t have arms, legs, a spine or a brain). Is there even a way to test if that premise is feasible?
This has nothing to do with molecular phylogeny either.
With respect, JP, you have left the subject of molecular phylogenies behind, more than once. Talking about something other than molecular phylogeny doesn’t falsify molecular phylogeny, now, does it? Nor does confusing inheritance patterns with inferring phylogenies.
Mark
------------------
Occam's razor is not for shaving with.
™ Version 4.2