DanielBoyd

Hi Ken, you're right, I'm oversimplifying the biochemistry, but there isn't space here to be complete on such complexities. My point is (whether we're talking DNA, RNA or protein), we're still on the molecular level here, and need to transcend that to explain the higher levels of organisation that are what life is really about. Your last sentence suggests that we share common ground. This is exactly what I am saying: that an emergent outcome is not the result of specification. Some people seem to think I'm making a mountain out of a molehill here: of course there are regulatory processes and stuff which actually determine outcome, but that that is trivial. Because the genome makes the parts it's still in control. I think this is more than semantics, and that the conceptual framework (self-assembly vs design as formative explanations for observed structures) is fundamental to the discussion. If you see the cell (and organism) as an emergent outcome, how do you interpret the role of the genome in that context? Sorry, I thought I'd answered this question. The point is that self-assembly does not need or use information to create its product. A snowflake does not need some design to create its filligree form: this is simply the result of water molecules interacting. The thermodynamics of this process allow something complex to arise out of something simple (chilly water vapour) without any design information or external mechanism to build it. Proteins are different from snowflakes in this sense. They have a design (the gene) and a construction mechanism (gene transcription). My contention is that the cell and the organism are like the snowflake, not the gene. They spontaneously grow without any external source of information to guide them out of the components (genetically-defined proteins, lipids, carbohydrates etc) that are available. We don't see cells or organisms 'crystallising' out of a random mix of molecules or cells like we do with snowflakes: instead growth and formation take place at the same time. But we do see something similar happening in the butterfly crysalis: All of the caterpillar's internal structures are first broken down into a disordered mush, which the re-forms into an entirely different ordered body. This is a classic case of self-assembly. Having said this, you are right that there is an 'information gap'. In terms of information theory the entropy of a cell/organism is lower than that of a random collection of molecules/cells, You can see this as: "it requires more information to define it". So where this information come from, if not from the genome, is a good question. In thermodynamic terms, this extra information (order) is created by the self-assembly process itself. It creates this order through the metabolic use of external energy sources. So the extra information is not stored but created. Does this answer your question?

It's a shame you should feel that way. I must say that I share the feeling: that you are failing to respond to the arguments I present concerning the limited role of the genome, simply repeating your own viewpoint. If I step back and try to look at the thread objectively, I see us both responding to what the other is saying, but maybe our mindsets are so far apart that bridging the distance from either side is nigh impossible. At the same time, I have the feeling that there is common ground that we are failing to find. At least we agree that genes determine proteins, that proteins are the most significant types of molecules in the cell, and that which proteins are present will therefore strongly influence the processes (both formative and functional) taking place in the cell. So fas so good?

Thanks for the explanation. I'm new to this forum and was expecting, since its a science forum, that I wouldn't need to explain that I'm not some religious zealot in disguise. Are there really such types in abundance here who aren't here for the science but to undermine it?

As long as I read things like "the genome is able to exploit the fact that proteins do in fact fold in predictable ways" I think that the man is not entirely made of straw. I think a more accurate formulation would be "The cell is able to exploit the fact that it has a gene at its disposal that creates proteins that fold in predictable ways". If you can comply with that, then maybe the man is made of straw and I'm just stating the obvious.

I very explicity indicate what I propose as this 'something else', and that is not supernatural. It is the all-too natural processes of self-assembly that shape most of the natural world. Perhaps my article was too long to read thoroughly (mea culpa) but I make this very clear. In the second paragraph I state that there are just two ways of creating things: design and self-assembly. God is not on the list. I also explain why design is not a systemic option for living systems: divine, genetic or otherwise. Any creationist should therefore be even more disgruntled than you ;-)

"So why are you looking for something else or something more? Perhaps it is simply that ‘self-assembly’ is more varied and pervasive and powerful than you currently perceive." I am not looking for anything more than self-assembly, and I certainly don't underestimate its power, quite the contrary, that amazes me too! The explicit proposition of the article is that it is nothing more and nothing less than highly advanced forms of self-assembly that creates biological structure. That is why I am arguing against the absolute, reductionistic, causative role of the genome. You say that DNA provides the self for the self-organising processes. If we ignore the non-proteinaceous components for a moment, this is true. But in an emergent, self-organising system knowledge of parts is not sufficient to determine the outcome. Living systems necessarily exert downward causation, in which the whole determines the behaviour of the parts. When they do, the parts loose control over the outcome. To take a different example, millions of termites work together to build termite mounds. This is not because there is some hidden or even implicit design of a mound in every termite (let alone in the termite's DNA!), Instead, it is the colony that determines the behaviour of the individual termite, resulting in it contributing to building the mound.

I have been a staunch atheist all my life, so it completely flummoxes me to be accused of religious motivation. I really am curious: what text in the original article leads you to this conclusion?

Reading all your comments, it strikes me that we seem to agree on the fact that genes only directly determine protein amino acid sequences, and that the rest is self-assembly or self-organisation. Where we seem to differ is in the significance we ascribe to what happens between the polypeptide sequence and the organism. You seem to be saying that by defining the protein components, DNA is completely responsible and in control of everything that happens in all the levels of organisation that are built above the molecular. In essence, this is an ultimately reductionistic explanation of life: define the components and you have defined life. This is to ignore the principles of emergence, which tell us that the whole can be more than the sum of the parts, and the parts therefore cannot explain the whole. The discussion of information quantity is important in this context, since emergence results in the existence of structures (information) at higher levels of organisation that cannot be derived from the parts. This goes far beyond the trivialisation of "sure, then they go on and do what they were designed to do". Living systems could not exist or survive without the kinds of powerful emergent top-down control that have been discovered in the course of the evolutionary process. This is acknowledged in homoeostatic systems: for instance our thermoregulatory processes determining whether brown adipose tissue cells are switched on to generate heat. However, before life can start controlling things it needs to create control systems. I see less evidence of appreciation of the determinative role of emergent top-down control in this even greater challenge. For instance, understanding how the brain works is pretty tough. Explaining how the brain (which has been quantified as the most complex structure in the known universe) comes to exist is next level. As I say in my opening sentence, many people see DNA as complete and sufficient explanation for the organism. Yet if any degree of emergence is involved then this is simply not true - and if there's one thing evolution is good at it's building powerfully emergent systems. That is why I wrote this article, and this discussion seems to confirm the necessity. Here, too, I hear people voicing the reductionistic explanation based on the sufficiency of genetic information, and trivialising the determinative emergent processes that actually create the massively improbable living structures we see. Clearly, my article spectacularly failed to transfer this meme! But your comments certainly help me underrstand why, so thanks for that!

Your assumption seems to be that any system must be the product of the expression of some source of stored external information. Therefore, if I say the genome does not contain enough information I need to provide an additional source. This assumption, however, is false. No self-assembling system is defined by external information: that is the very definition of self-assembly. There is no instruction for building a snowflake, and there is no instruction for building an eyeball. Both are the result of inherent interactions between their component parts. So no need to hold your breath ;-) I'm not going to tell you where this information is stored, because it doesn't exist. Instead, self-assembling systems create their own information (thermodynamically: use external sources of energy to decrease their entropy). So you would agree with me that the morphogenesis is a form of self-organisation, and therefore not genetically defined? After all, the definition of self-organisation is that it is not defined by some external influence or instruction, but arises directly through the interactions between components. I agree (almost) entirely. To be a bit more exact about it: DNA knows how to build polypeptides, which form the basis for the self-assembly of proteins through the creation of secondary and tertiary structures, which constitute a significant part of the collections of molecules that self-organise to form organelles, with self-organise to form cells, which self-organise to form tissues, which self-organise to form organs, which self-organise to form organisms. You proposition is that nothing more than the DNA sequence is necessary to define and determine the outcome of all these self-organisational processes?

Thanks for bringing in the serious science, Arete. What I am saying is actually supplementary rather than contrary to what you write. Certainly developmental genetics is an important factor in morphological development. But does this equate to sufficient causative explanation for the existence of complex morphological structures? The products of Hox genes are transcription factors that influence the expression of other genes. In other words, they determine how much of some other protein should be produced by the cell. This secondary protein then goes on to play a role in cellular or intercellular processes, along with a whole load of other molecular components. The Hox gene does not define the secondary protein that is produced (that is the task of the gene it activates), nor the way in which it all works together in an organised fashion to determine how the cell functions in its environment, let alone how millions of cells with different patterns of gene expression work together to create a morphological structure. All it does is say: let's send this much of this protein into the mix. The outcome of any self-organising process will obviously be determined by the participating components. In so far as a Hox gene influences how much of a particular component is present, it will therefore have an effect on the outcome. However, this is most certainly not the same as saying that the Hox gene (or even all of the Hox genes together) provides sufficient determinative explanation for the outcome. Furthermore, we have the question of the regulation of the expression of Hox genes: who directs the director? Hox genes cannot cause their own expression, so even they are only players in the greater scheme of things. To explain such a small part of this immensely complex system and deem this as sufficient explanation for the whole is surely to overstate your claim.

If only! Typing "A gene for" into Google Scholar produces 245,000 hits. Admittedly, many of these are biochemical studies where there is a direct relationship, but certainly not all. Within the scientific community there is an implicit or explicit assumption that hereditability of traits is synonymous with genetic determinism. Therefore anything that is inherited must be genetically determined. Therefore morphology and physiology are genetically determined. That is the assumption I am challenging with this article.

Great, now we are getting to the meaty stuff! Certainly gene activation and transcription is a more complex mechanism than just sticking an amino acid to a codon, with intermediate steps involving RNA. And, absolutely, there are feedback mechanisms involved in activation. But if we look at how these actually work, things get trickier. Hox genes have an important role in organismal development and are found across the animal kingdom. However, the same genes have different effects in different species https://www.sciencedirect.com/science/article/pii/S0960982206003216 . Does this not show that they are participants in the organisational processes they take part in, rather than determining it's specific outcome, as would be the case if they constituted a determinative design?

Sorry, people, I come back from my day job and find a whole thread has grown that I have failed to respond to. Gotta earn a living! I think the general conclusion is that we don't actually disagree that much: the only direct information present in the genome is for the amino acid sequence, but the amino acid sequences determine how the proteins produced behave in the cell. In this way genetic differences indirectly exert some control over the formation and function of cells, and more indirectly on the interactions between cells that lead to the formation of multicellular organisms. My point - that the genome does not literally contain a design for these things - may seem trivial, but it's surprising how often you still read about 'a gene for this' or 'a gene for that', when in fact what is happening is that a genetic/protein variation is only having an indirect effect on the complex, emergent, self-organizing system it is a part of. A logic similar to putting the wrong size spark plug in your car and claiming to have explained how the engine works when it splutters. This is why the discussion of the quantity of information in the genome is relevant. In information theoretical terms, the emergent processes taking place create more information than is present in the genome. This is not just a 'decompression' of hidden content in the genome, or a random expansion, but a highly specific generation of information that is simply not present there. This is where the 'missing information' comes from. Re-reading my article, and your responses to it, makes me realise not so much that I am wrong or you are wrong, but that I have spectacularly failed to define the issue sufficiently clearly. So now we're like the two knights meeting at a crossroads arguing about whether the statue is silver or gold (if you know the story). This could partly be explained by the fact that the article was originally longer, more complete and more formal, but I reduced it down and made it a little more playful in the hope of at least getting it read. I think this particular reduction in information didn't help: the emergent processes that followed created a bit of a monster! What the solution is I'm not sure, but I would like to thank particularly Eise, John, SharonY, Ken and Dag1 for your feedback (sorry that I didn't get round to responding to your extensive post on the first page).

This is quite wonderful: the examples of cellular processes and structures being presented all demonstrated the 'incompleteness' of genetic information as a determining explanation for the functional cell. With respect to Brownian motion, certainly this may (partly) solve the 'transport problem' of how to move proteins from where they are made to some other place in the cell, but the cell is far from a mixed bag of molecules all zipping around randomly and bumping into each other. It contains many stable structures (usually built out of lipid membranes with specific proteins embedded in them). So getting (and keeping) proteins and other molecules in the right place is essential (and, of course, not genetically determined). It may have occured to you that up to now, we are still inside the cell: the molecular environment in which the genome still plays a fairly direct and dominant role. Even here, we are encountering the importance of self-assembly in addition to genetic information to get things done. The original article, however, is primarily about the (massive!) step to the next level of organisation: that of the cellular differentiation and organisation involved in creating a multicellular organism. What are your ideas on the relationship between genetic information and. say, the structure of the kidney, of the layers of cells that make up skin tissue?

And that is a compliment (in my book)