Deterministic versus stochastic mechanisms
So cells seem to baffle us constantly with their concise and efficient machinery. Cells make decisions about when to divide and what have you based on complex networks of signaling effectors and receptors. These processes are considered deterministic when A leads all the way down to Z, and there's some very concerted feedback loop, and everything just has to work out that way and they cell is going to divide goddamn it! If it doesn't, that means something is wrong - genetically. On any normal day, these decisions are hard coded in the genome.
Or is it, though? Yes and no, right? Almost every cellular mechanism seems to be pretty damn deterministic and from science's perspective, it's just a matter of teasing out all the molecules involved and their relationships. Also, in a multicellular organism, it's easy to think of cellular decisions as being deterministic because, hey, they just got a signaling molecules or hormone or whatever and they're just doing what they're told. If not, it's pathology, and that's just another deterministic mechanism. Even cancerous developments, which occur from random mutagenic events, are understood as rather deterministic as a function of specific oncogenes and tumor suppressor genes being mutated. If they got the x-ray that hit those genes, that's not their fault, it's the stochasticity of the world around them.
So like, what they hell? Biochemistry is random. Things bump into each other all A+B -> C -style. The transport of molecules around the cytosol is governed by random diffusive motion. But it's not really. There is active transport, molecular crowding, and all that mess. The construction of the cytosol is "designed" encourage associations of molecules that are supposed to, and discourage others. Biophysical constraints within the cell reign in the random diffusive kinetics that govern simple chemical reactions.
Still - why? The cell is not a person, you know, just deciding what's best for itself. Sure, cells evolve to do will for themselves - impressively so, don't get me wrong - but they can't just eliminate randomness entirely. They can't escape from the inherently stochastic nature of the physical world, and therefore neither can they construct everything in terms of deterministic switches.
From a signals perspective, biological networks can be shockingly robust. Feedback loops, branch points, shunts, scale-free networks, whatever. But in all senses of the word, single cells are not sentient. They can't make conscious decisions. Deterministic switches within them are actuated by environmental stimula that enact carefully conserved and enforced networks of response, achieving fitness. Bam. Is it that simple that everything just responds like clockwork to everything else like some crazy deistic machina? No.
I mean, look at the beginning of life. There was no deterministic mechanism there, certainly. We don't really know what the hell happened, but we are pretty sure it had to do with a big bucket of chemical species interconverting, getting lumped up in oily bubbles and then things just got out of hand you know? Sooner or later you had nucleic acids and proteins and things were crazy. That is the essence of stochasticity, amirite?
OK so this is starting to get unfocused. Let's look to a more specific examples, and I apologize that it has to do with microbes and not fancy humans or mice or whatever. I think microbes are interesting, and they certainly have a lot to teach us as a model organism. Anyway, on to the "example" ...
Let's say you are a bacterium. Like, e coli, for example. And you have a whole bunch of brothers and sisters like ~1 million or whatever. You all share the same genotype, because, duh, you're all just clones of eachother. You don't have meiosis or sexual reproduction, you stupid fuck, you're not that high up the food chain. Anyway, you and your clone family should all be exactly the same, right? No - why? well you wouldn't want that, that would be stupid because we all know that variety is the spice of life. Fitness is predicated on some kind of diversity.
Now here's the problem: how do you have phenotypic individuality while having genotypic homogeneity? Traditional evolution - e.g. spontaneous mutations in the genome resulting in increased fitness - is WAY to slow for most real world problems in the life of a bacterium.
This is where I start copying wholesale out of a review I just read in a microbiology journal about this topic. In any case the answer is to embrace the noise. This time, instead of minimizing the random nature of chemical reactions, you want to use that to your advantage. There are A TON of examples of this and people are starting to get hip to this and finding more. The simplest way this usually works in through intrinsic biological noise, or the copy-number effect.
Intrinsic biological noise applies to very low-copy proteins/mRNAs. If you only make a handful of mRNA transcripts for a given gene, natural kinetic fluctuations (biochemistry is random) that add or subtract one or two transcripts will have a much larger effect than on highly transcribed genes. If those mRNA are then translated into proteins at a very high rate, then the noise is amplified. As a result, certain genes are under control of a high degree of intrinsic of biological noise, leading to these certain genes having largely differential expression across individuals in a genotypically homogeneous population.
So there it is. Stochastic mechanisms lead cells to have highly individual phenotypes and this leads to specific behaviors and cellular decisions. In these case, the are hardwired to use random noise to make cellular decisions, rather than deterministic switches that respond accurately to specific inputs.
Like I said before, this is not new (Well, it's kind of new - it's certainly a 21st century thing). What I'm saying is largely backed up on personal conversations with professors and examples described in various review articles. Now I'm thinking about where to go from here.
Now, I think this is pretty damn cool, but what about the people that don't want to hear about our single-celled friends? It seems like higher-order eukaryotic cells battle down the noise pretty hardcore. I mean the fact that intracellular cargo is actively moved around the cell by protein motors that convert chemical energy into mechanical work is just one of many examples of how cells fight chemical randomness. If they want the protein in the nucleus, they give it a nuclear localization sequence. If the want it out, they give it a nuclear export signal. you get the idea.
Is there room for randomness in the mammalian cell? Or is all squashed out? I'd like to read a review about that. It seems like it's not a thing. Randomness NOT being a thing in mammalian cells is interesting, but I would also be find out where it is lurking. I think my friend is right that epigenetic regulation is the big stochastic hotspot, but I think there's more.
Anyway, for me it tends to come back the "intelligence" of cells. It makes me ask, why? I always thought it was an orchestration of a seemingly infinite number of deterministic mechanisms. Noise exists in biology though, and it has to figure into the picture somehow. Some cells use this noise to make decisions, but for the most part, the "really interesting cells" (our cells) don't seem to. Rather, they seem to have as many safe guards against noise as you can, in the form of redundancy, alternative pathways, and feedback loops.
I guess the only way to proceed from here is sort of more of the same? Try to dissect profound but mysterious aspects of cellular machinery in hopes to find out more about them. Somehow, that seems less edgy than I was hoping this rant was leading me. Haha, I'll end this now.
Or is it, though? Yes and no, right? Almost every cellular mechanism seems to be pretty damn deterministic and from science's perspective, it's just a matter of teasing out all the molecules involved and their relationships. Also, in a multicellular organism, it's easy to think of cellular decisions as being deterministic because, hey, they just got a signaling molecules or hormone or whatever and they're just doing what they're told. If not, it's pathology, and that's just another deterministic mechanism. Even cancerous developments, which occur from random mutagenic events, are understood as rather deterministic as a function of specific oncogenes and tumor suppressor genes being mutated. If they got the x-ray that hit those genes, that's not their fault, it's the stochasticity of the world around them.
So like, what they hell? Biochemistry is random. Things bump into each other all A+B -> C -style. The transport of molecules around the cytosol is governed by random diffusive motion. But it's not really. There is active transport, molecular crowding, and all that mess. The construction of the cytosol is "designed" encourage associations of molecules that are supposed to, and discourage others. Biophysical constraints within the cell reign in the random diffusive kinetics that govern simple chemical reactions.
Still - why? The cell is not a person, you know, just deciding what's best for itself. Sure, cells evolve to do will for themselves - impressively so, don't get me wrong - but they can't just eliminate randomness entirely. They can't escape from the inherently stochastic nature of the physical world, and therefore neither can they construct everything in terms of deterministic switches.
From a signals perspective, biological networks can be shockingly robust. Feedback loops, branch points, shunts, scale-free networks, whatever. But in all senses of the word, single cells are not sentient. They can't make conscious decisions. Deterministic switches within them are actuated by environmental stimula that enact carefully conserved and enforced networks of response, achieving fitness. Bam. Is it that simple that everything just responds like clockwork to everything else like some crazy deistic machina? No.
I mean, look at the beginning of life. There was no deterministic mechanism there, certainly. We don't really know what the hell happened, but we are pretty sure it had to do with a big bucket of chemical species interconverting, getting lumped up in oily bubbles and then things just got out of hand you know? Sooner or later you had nucleic acids and proteins and things were crazy. That is the essence of stochasticity, amirite?
OK so this is starting to get unfocused. Let's look to a more specific examples, and I apologize that it has to do with microbes and not fancy humans or mice or whatever. I think microbes are interesting, and they certainly have a lot to teach us as a model organism. Anyway, on to the "example" ...
Let's say you are a bacterium. Like, e coli, for example. And you have a whole bunch of brothers and sisters like ~1 million or whatever. You all share the same genotype, because, duh, you're all just clones of eachother. You don't have meiosis or sexual reproduction, you stupid fuck, you're not that high up the food chain. Anyway, you and your clone family should all be exactly the same, right? No - why? well you wouldn't want that, that would be stupid because we all know that variety is the spice of life. Fitness is predicated on some kind of diversity.
Now here's the problem: how do you have phenotypic individuality while having genotypic homogeneity? Traditional evolution - e.g. spontaneous mutations in the genome resulting in increased fitness - is WAY to slow for most real world problems in the life of a bacterium.
This is where I start copying wholesale out of a review I just read in a microbiology journal about this topic. In any case the answer is to embrace the noise. This time, instead of minimizing the random nature of chemical reactions, you want to use that to your advantage. There are A TON of examples of this and people are starting to get hip to this and finding more. The simplest way this usually works in through intrinsic biological noise, or the copy-number effect.
Intrinsic biological noise applies to very low-copy proteins/mRNAs. If you only make a handful of mRNA transcripts for a given gene, natural kinetic fluctuations (biochemistry is random) that add or subtract one or two transcripts will have a much larger effect than on highly transcribed genes. If those mRNA are then translated into proteins at a very high rate, then the noise is amplified. As a result, certain genes are under control of a high degree of intrinsic of biological noise, leading to these certain genes having largely differential expression across individuals in a genotypically homogeneous population.
So there it is. Stochastic mechanisms lead cells to have highly individual phenotypes and this leads to specific behaviors and cellular decisions. In these case, the are hardwired to use random noise to make cellular decisions, rather than deterministic switches that respond accurately to specific inputs.
Like I said before, this is not new (Well, it's kind of new - it's certainly a 21st century thing). What I'm saying is largely backed up on personal conversations with professors and examples described in various review articles. Now I'm thinking about where to go from here.
Now, I think this is pretty damn cool, but what about the people that don't want to hear about our single-celled friends? It seems like higher-order eukaryotic cells battle down the noise pretty hardcore. I mean the fact that intracellular cargo is actively moved around the cell by protein motors that convert chemical energy into mechanical work is just one of many examples of how cells fight chemical randomness. If they want the protein in the nucleus, they give it a nuclear localization sequence. If the want it out, they give it a nuclear export signal. you get the idea.
Is there room for randomness in the mammalian cell? Or is all squashed out? I'd like to read a review about that. It seems like it's not a thing. Randomness NOT being a thing in mammalian cells is interesting, but I would also be find out where it is lurking. I think my friend is right that epigenetic regulation is the big stochastic hotspot, but I think there's more.
Anyway, for me it tends to come back the "intelligence" of cells. It makes me ask, why? I always thought it was an orchestration of a seemingly infinite number of deterministic mechanisms. Noise exists in biology though, and it has to figure into the picture somehow. Some cells use this noise to make decisions, but for the most part, the "really interesting cells" (our cells) don't seem to. Rather, they seem to have as many safe guards against noise as you can, in the form of redundancy, alternative pathways, and feedback loops.
I guess the only way to proceed from here is sort of more of the same? Try to dissect profound but mysterious aspects of cellular machinery in hopes to find out more about them. Somehow, that seems less edgy than I was hoping this rant was leading me. Haha, I'll end this now.