I think that you are using different senses of "predict" here.
We don't know that there isn't a minimal gene set that would enable evolution into something of equivalent complexity to modern animals and plants in the time available. It only seems likely that this is true.
But if we accept that the LUCA is a community of simple organisms sharing genes it is vanishingly unlikely that evolution would only have a minimal gene set to work with.
So, on this basis your "not a prediction" seems to be a stronger statement than your "prediction", which simply can't be the case if you are using a consistent standard.
I would also comment on this:
quote: This is because, under non-teleological evolution, the LUCA could have been simply a self-replicating molecule, encoding only a few genes.
Even taken as a vague idea, ignoring what we know about the development of early life this seems false. How can a simple self-replicating chemical have genes ? If we take into account the fact that DNA seems to be a relative latecomer (and ask yourself why we would have evidence pointing to this if life on Earth started with artificial prokaryotes) any LUCA you could find by investigating the homology of genes would necessarily be the product of considerable evolution, with DNA and a gene set already in place - you can't go back to anything earlier.
quote: Would you care to elaborate on this? With a LUCA that has a minimal gene set, there's really no reason to suppose that complex life forms could arise.
You're looking at the wrong question. The real question is whether it is possible to engineer a minimal gene set which would encourage the formation of complex life. Given the size of the minimal gene set that you suggest, it seems that that is at least something that could plausibly be true.
quote: You'd have to guarantee the rise of mitochondria before you can get complex life forms
That would seem pretty difficult to do by front-loading a single organism. Wouldn't it make a lot more sense to engineer eukaryotes with artificial mitochondria - if the engineers chose the symbiotic route at all?
quote: What does horizontal gene transfer have to do with a minimal gene set? I.e., sure, this LUCA community could be transferring genes, but these genes would be part of the minimal gene set.
If the LUCA is a population of organisms exchanging genes it would be expected to have a greater diversity of genes than if it were a single individual organism. And thus far more likely to have more than a minimal genome (in aggregate).
quote: A self-replicating RNA molecule could specify a protein sequence.
A notion that only makes sense if we assume a mechanism that can "read" that specification and produce a protein. Which means that we are talking about something more complex than a simple self-replicating molecule
quote: I haven't found any convincing evidence that DNA evolved from RNA. I am aware of a number of papers on the subject, but I'll (hopefully!) discuss this evidence in another thread (so we don't get sidetracked here).
For the purposes of this discussion it is going to be better if you at least take the leading views of the origin of DNA as possibilities that have to be considered. To the best of my knowledge, the RNA World hypothesis is widely accepted among researchers into the origins of life.
quote: Yes, but under the FLH, the LUCA isn't the product of considerable evolution. Maybe I'm not getting your point with this statement?
Of course, I was talking about the prevailing evolutionary view. Not that FLH makes any prediction on the matter at all.
quote: The minimal genome would be around 250 genes. You're basically suggesting that the designers take calmodulin and fashion it such that it is necessary for the prokaryotic cell plan, are you not?
Certainly not! I'm suggesting that it is plausible that there is a set of 250 genes that would encourage the evolution of complex life.
quote: Eukaryotes aren't very good terraformers compared to prokaryotes, and FL is intimately linked to terraforming. If you engineered the first bunch of cells in the right way, the origin of eukaryotes becomes quite likely. Thus we see that multicellularity has evolved independently several times, pointing to the plausibility of the evolution of multicellular life forms, given specific initial conditions.
This seems to make it more likely that if FLH were true, that the designers would start with something like eukaryotes, at least in capabilities rather than try to arrange for a suitable endosymbiosis event to happen eventually.
quote: Yes, but the LUCA could also be a self-replicating RNA molecule, with - as you correctly point out - small amounts of molecular machinery needed for the formation of primitive proteins.
I don't see that as very plausible, unless you assume that the LUCA was engineered.
quote: Yes, but if I accept the RNA world hypothesis - that is, the thesis that life evolved from compounds on earth - my front-loading hypothesis is blown out of the water, effectively. So, when you suggest that I take the leading view that DNA evolved from RNA, you're essentially saying that I should take the leading view that teleology has not played a role in the history of life on earth.
I am not asking you to believe in the RNA world. I am saying that you have to accept it as an important part of a serious opposing position. If you don't do that then you are committing something akin to the strawman fallacy.
quote: Yes, if this set of 250 genes encoded stuff like calmodulins, histones, tubulin, dynein, etc. But such a set of genes wouldn't be something that is required by all life forms
So long as it encourages the development of complex life forms, it doesn't matter what the genes code for. They need not be analogs of proteins used by known life.
quote: And this is where we get our FL prediction: at minimum, the LUCA would have consisted of 250 genes (actually, fewer, given that LUCA didn't have to have a complex prokaryotic cell organization). But a minimum gene set consisting of the genes absolutely required for life doesn't allow you to front-load anything.
That is the very assertion that I am disagreeing with. Given 250 distinct proteins it is simply not clear that they would be insufficient for front loading, given the modest goals assumed.
quote: Under the RNA world hypothesis, how did the first proteins originate? They would have originated long before the origin of the LUCA, and prior to the LUCA you'd have very simple cells (under the non-telic model) - possibly consisting of primitive membranes such as lipid bubbles - and these cells wouldn't require 250 genes.
My understanding is that more complex RNA life pioneered the use of proteins. And, of course, this is before the LUCA. Obviously proteins useful to this RNA life would tend to be preserved into the era of early DNA life, even if they were not absolutely essential. It seems very unlikely to me that there would be no such proteins, save for those absolutely necessary to cellular life.
quote: True, but proteins that encourage the development of multicellularity and Metazoa etc. aren't likely to be found in a minimal gene set, because such genes only encode the bare necessity, which means there'd be no "front-loading." There's a difference between simply designing the first life forms and specifically engineering it such that it front-loads the appearance of complex life forms.
That's the assumption that I am challenging. Given that we know that proteins may have multiple uses, plus the possibility of near variants having still more uses 250 different proteins should give you quite a range of options to encourage the development of complex life.
quote: Yes, but you're forgetting that these 250 distinct genes would also have to be part of a minimal gene set.
No, I'm not.
quote: So, you'd need to get one of the genes necessary say, for the Krebs cycle, and engineer it such that it actually does nudge evolution in the direction of complex life forms. Or you could take SecY, which AFAIK is a necessary protein for life, and engineer it so that it biases evolution in favor of multicellular life. But then, once you do that, there's no reason to suspect that we'd have something that looks like SecY anymore. And, unless you have evidence suggesting otherwise, there's nothing inherent in SecY that would bias evolution towards the development of complex life forms, is there?
I'm not sure that I understand your argument here. Obviously we can't require that the front-loaded gene set would necessarily do things the way earthly life does them - that would be begging the question. So obviously we don't have to have something looking like SecY nor do we have to have the precursors to the Krebs cycle. All we need is a set of proteins that will work as a minimal life form while strongly encouraging the development of more complex life forms. You say that it can't be done, not even by your hypothetical designers, with their unknown capabilities. BUt I'm still waiting to see a real reason.
quote: Actually, PaulK, the burden of proof is on you to provide evidence for the assertion that one can design a gene set that both functions as part of the minimum genome and also encourages the development of Metazoa etc. (incidentally, such a strategy for front-loading would be a pretty clumsy design since with only 250 genes you'd have to front-load the appearance of a whole bunch of proteins necessary for the rise of eukaryotes and Metazoa).
No, I don't have to support assertions that I haven't actually made.
But to support the point I DID make:
The hypothetical designers would be free to choose any set of proteins that would work. They would, for instance be completely free to choose completely unrelated proteins. They would NOT be limited to using the proteins used by Earthly life.
Proteins are versatile and can be and are co-opted for different uses.
Mutation may enable other functions in proteins.
250 proteins is quite a large number.
Taking these considerations into account it seems to me that the hypothetical designers could do a far better job of enabling the development of complex organisms using a simple 250 gene single-celled organism than evolution could. (Which does raise the question of how you could consider the evolution of modern life plausible at all - according to you it would have to start with something greatly inferior to something you consider inadequate).
But let us get to the real point. f
For your argument to be truthful you need a consistent standard for "prediction".
That is you need a satisfactory standard that YOUR views satisfy, while the arguments from the evolutionary side do not.
That is rather difficult when your view is just an assumption that seems intuitive.
So, do you have a stronger argument that actually deals with the issues, or are you just going to evade the problem ?
quote: Quite right. But they have to include the 250 or so genes necessary for carbon-based life forms to exist. And they'd add extra genes to the original genomes so that front-loading could follow.
Again, you are just repeating your assumption.
quote: Okay, let me get your position right: are you suggesting that the designers could engineer the 250 genes needed for life such that they also favor the trajectory of evolution in the direction of Metazoa etc.?
I am suggesting that it is possible that they might be able to engineer the proteins to favour an evolutionary trajectory that would produce the features that they were interested in, yes. I don't accept that those possibilities are only satisfied by the metazoa as we know them.
quote: Not sure what you mean by all of the above. Care to elaborate?
Your standard for prediction seems to require absolute certainty, because you reject near-certainties as predictions when they come from the evolutionary side. Obviously to meet this standard you need a very strong argument that a gene set of 250 proteins is inadequate even with the advantages of engineering.
I don't see any such argument from you at all. I don't see anything more than an opinion supported only by intuition. Indeed you can't even say how you could front-load the introduction of mitochondria to a standard you would consider adequate, nor offer any reason why the proposed designers might not start off by including something equivalent to basal eukaryotes in their biological package. Or indeed, finding some other solution to the engineering problem that mitochondria solve.
quote: It's not exactly an assumption IMHO. It's an argument based on what we know about biology: life requires a fairly specific core set of genes and there is little evidence that those genes, in themselves, favor the appearance of complex life forms.
It would be more accurate to say that Earthly life requires a fairly specific set of genes, and there's little evidence that THOSE genes favour the emergence of complex life forms (although that's more to do with our current ignorance than our knowledge). When we start talking about hypothetical designed life then we're talking about something very different.
quote: Where is the evidence that one can engineer a gene set that not only fills the roles necessary for the existence of life but also favor the appearance of complex life forms?
Where's the evidence that it can't be done ? Remember that your requirement for a prediction is absolute certainty - an opinion won't cut it. I've given reasons to think that it might be possible and you haven't really answered them. That's not good enough to have a solid prediction even by my standards, let alone yours.
quote: I have no idea where you got that idea.
Yes, you apologetic types often have problems remembering what you said:
...one scenario that is perfectly acceptable under Darwinian evolution is that the LUCA had only a minimal genome. Such a scenario is not, however, compatible with FLE.
According to you the fact that non-telic evolution doesn't absolutely rule out the possibility of a LUCA with a minimal gene set is sufficient to say that it is not a prediction. Therefore FLE must absolutely rule out the possibility that a minimal gene set could be constructed that would provide sufficient guidance for the purposes of it's inventors. Now, without any idea of the limits of design or any clear idea of how that guidance might work or even the purposes of the designers you aren't in a position to do that.
So lets get rid of the double standard. At present there's a stronger case for a LUCA with a non-minimal gene set under the assumption of non-telic evolution than there is under the FLE. That means that it isn't a prediction of FLE that can separate it from standard evolutionary views.
quote: However, if we follow the line of thought that Metazoa were, in fact, the intended outcome of front-loading, from here we can make real predictions that are not made by the non-teleological model. The non-telic view of life does not require that prokaryotes have ubiquitin homologs.
I disagree. If ubiquitin is essential to eukaryotic life, and if eukaryotes and prokaryotes have a common ancestor then non-telic evolution predicts that at least some prokaryotes should have a ubiquitin homolog.
The reasoning as follows:
If eukaryotes are descended from prokaryotes then ubiquitin must have evolved in the prokaryotic ancestors of eukaryotes (because essential proteins don't just appear at the exact moment that they are needed)
If prokaryotes are descended from eukaryotes then ubiquitin must have been present in the ancestors of all prokaryotes and lost through evolution, after the divide between the two Kingdoms.
quote: Non-teleological evolution does not predict that ubiquitin will have a prokaryotic homolog because the ubiquitin gene could easily have been pieced together from different pieces of DNA, in much the same way that T-urf13 evolved.
That doesn't affect my reasoning, though. HOW ubiquitin appeared is not an issue. WHEN it appeared in the evolutionary history of eukaryotes and prokaryotes is the issue. And my argument covers that.
quote: In such a scenario, given the deep-time involved and the fact that this protein would have been cobbled together from short stretches of DNA, we would almost certainly not be able to trace this homology in prokaryotes - and, of course, structural analyses wouldn't reveal any homologs of ubiquitin in prokaryotes.
This fails to address my argument. If ubiquitin is truly essential for eukaryotes then this patching together almost certainly appeared in a prokaryote - the alternative is vanishingly unlikely. As I said, genes don't magically appear in the first organism that needs them. Therefore there must have been prokaryotes with ubiquitin even in your scenario. So why wouldn't we expect to find homologues ?
quote: Okay, here's a simple true/false question: If the Metazoa we see today was the intended outcome of a front-loading scenario, could we make testable predictions from this premise?
I don't think so. I would add that that hypothesis assumes too much - it would be much better to construct a plausible scenario and make predictions from that.
quote: No, non-teleological models do not predict that crucial eukaryotic genes will share deep homology with functional but unnecessary (for life) prokaryotic proteins. I've explained why many times, but here it is again: under the non-telic model, it is completely reasonable for the LUCA to have no more than a minimal genome. I have supported this contention with references to the scientific literature.
Firstly we note that neither you nor your sources present any reasoning to support this assumption.
Secondly, we must note that evolution is not a goal-directed process and as such it seems somewhat unlikely that the first cell would have acquired only those genes absolutely necessary to form a cell.
Thirdly we wouldn't expect the LUCA to be the first cell. And it's even less reasonable to expect there to be no ancestors shared between eukaryotes and some prokaryote since the LUCA, which could also produce deep homologies.
quote: ...This indicates that in contrast to early hypotheses, LUCA was far from being a minimal cell because its genome was far from a minimal genome." (emphasis added; from "Origins and Evolution of Life: An Astrobiological Perspective," Muriel Gargaud, Purificación López-Garcìa, Hervè Martin, Cambridge University Press, 2011)
Read that carefully, then tell me that it's not reasonable, under non-teleological models, for the LUCA to have only a minimal genome and be only a minimal cell.
ALright. It's not reasonable, under non-teleological models, for the LUCA to have only a minimal genome and be only a minimal cell.
Seriously all you have is a claim about "early hypotheses" with no indication of how they were arrived at. Where's the reasoning that would make them reasonable ?
The issue here really isn't whether the non-telic view of life can potentially explain the observation that crucial eukaryotic proteins share deep homology with functional but unnecessary prokaryotic proteins. The real issue is whether the non-teleological model predicts this.
And whether FLE predicts it. That's another of your claims that is somewhat lacking in support.
So *IF* it appeared only shortly before eukaryotes arrived it MIGHT have been lost. That doesn't mean that we don't expect to see homologues.
Of course if it was as useful to prokaryotes then that wouldn't be likely to happen. Of course we wouldn't expect to see only distantly related homologues in prokaryotes either. So ubiquitin hardly supports your ideas.
quote: Challenge: find a single paper in the scientific literature that argues that it is not reasonable under the current paradigm for the LUCA to have only a minimal genome.
I think that I have a paper that about qualifies - and I note that you have yet to find one paper that actually ARGUES otherwise rather than, for instance, stating that it was assumed.
Even if you judge otherwise this paper at least argues that a non-minimal genome is more reasonable and calls the reasonableness of the assumption of a minimal genome into question:
We think moreover that there are general lessons to be learned from evolutionary biology studied at different organismal levels. For example, discussing the Cambrian radiations of primitive animals, Balavoine and Adoutte (1998) stressed ‘the traditional intellectual bias for increasing complexity in evolution’ and Adoutte et al. (1999) concluded that ‘animal diversification… must have been triggered primarily by external factors acting on a “preadapted” (meaning possessing many features prone to further speciation or functional cooption), already genetically complex metazoan’ (our emphasis). The parallel with the problem discussed here is obvious. The fuzzy organization of a redundant genome inherited from the progenotic era, made of scattered domains or subdomains separated by interstitial sequences must have been a far better substrate for further evolution than a prokaryote-like genome, which would have had to acquire several capital novelties and constantly increase in complexity
quote: Meanwhile, there are a whole bunch of other studies that argue that the LUCA had a minimal genome, that it was a simple virus-like entity, that it had an RNA genome, that did not encode tRNA genes, etc. etc. All these scenarios are fully compatible with the non-teleological framework. They are not, however, consistent with the front-loading hypothesis, and this allows us to make testable predictions from that perspective.
Perhaps you should find some of these studies. You don't seem to have quoted any. (And I have to say that bare front-loading is also compatible with a large number of scenarios - you need to specify what the front-loaders intended before you can say what is front-loaded).
quote: Umm, if the original protein had been lost, then we wouldn't find any homologs. And since it's a plausible occurrence, we can't predict deep homology or lack of deep homology from a non-telic perspective.
It would be more likely to be retained than lost, therefore we have a reasonable expectation of finding homologues.
On the other hand the actual data is more consistent with a non-telic evolutionary view than it is with your front-loading.
Ubiquitin is highly conserved among eukaryotes, thus by your reasoning it must have been supplied with a mechanism to keep it adequately conserved. The only homologues known among prokaryotes are very distant, indicating that no such mechanism currently exists.
Moreover, some prokaryotes have systems similar to that where ubiquitin is used in eukaryotes, but do not use ubiquitin or homologues.
Thus we have neither a reason to believe that the importance of ubiquitin is anything more than a "frozen accident" or that it could have been supplied in a distant ancestor of eukaryotes.
quote: But from a non-teleological view, we have no way of knowing if it was terribly useful to prokaryotes. And useful genes get lost in lineages all the time. That's the Darwinian explanation for why we don't see non-flagellar homologs of, say, FlgD, FliD, FliL, and a whole bunch of other flagellar proteins (that, and saturation of informative sites in protein sequences).
Since your whole scenario rests on useful proteins being certain to be retained - and kept intact for functions other than the one they currently serve (a more difficult proposition still!) - this rather seems to suggest that your scenario is unworkable.