How Life Works

Prologue

• Looking to the genome for an account of ho life works is rather like looking to a dictionary to understand how literature works.
• The new picture dispels the long-standing idea that living systems must be regarded as machines.
• Living entities are generators of meaning. They mine their environment (including their own bodies) for things that have meaning for them: moisture, nutrients, warmth. It is not sentimental but simply following the same logic to say that, for we human organisms, another of those meaningful things is love.
• Life is a hierarchical process, and each level has its own rules and principles: there are those that apply to genes, and to proteins, to cells and tissues and body modules such as the immune system and the nervous system. All are essential: none can claim primacy.
• Genes don't generally specify unique outcomes at the level of cells and organisms.
• Recurring themes and principles:
  • Complexity and Redundancy
  • Modularity
  • Robustness
  • Canalization
  • Multilevel, Multidirectional, and Hierarchical Organization
  • Combinatorial Logic
  • Self-Organization in Dynamic Landscapes
  • Agency and Purpose
  • Causal Power

1. The End of the Machine: A New View of Life

• Living things are, you could say, those entities capable of attributing value in their environment, and thereby finding a point to the universe.
• Meaning-generators are successful entities in a Darwinian world. Making meaning is a great way of staying alive and propagating - so much so, indeed, that it's probably the only way to be alive at all.
• The point is that we need to acknowledge what evolution does to matter: it gives matter goals and functions. That is what makes evolved life so special.

2. Genes: What DNA Really Does

• Each of our somatic cells contains 46 chromosomes: two copies each of 23 different varieties. Other animals have different numbers: cats have 19 pairs, dogs 39 pairs.
  • The gametes (eggs and sperm) are special in that they each contain only one set of chromosomes. This makes them haploid cells, as opposed to the diploid somatic calls.
  • The single exception in our bodies are the red blood cells, which contain no DNA; they are simply packed with oxygen-ferrying hemoglobin proteins.
• Chromosomes are made of protein and of deoxyribonucleic acid (DNA), which is in turn a polymer of four types of chemical unit (nucleotides), containing a sugar molecule (deoxyribose), a phosphate group, and a substance belonging to the general class of molecules called bases (adenine (A), cytosine (C), guanine (G), and thymine (T).
• The unification of Darwin's theory of natural selection with the Mendelian inheritance of particulate genes gave 20th C biology its central explanatory framework, dubbed the Modern Synthesis in 1942 by Julian Huxley.
• The gene's eye view of life (indeed, even of evolution) is shaped by a particular scientific model and is valid only within the context of that model. It does not and cannot deliver an account of the world as we find it. The problem with atomizing organisms into genes is that genes are not alive - and once you have set aside life to get to the gene, you can't get it back again. The gene is far too atomized a unit to tell us much at all about how life works.
• Genes with names awarded in one context turned out to be identical to genes given different names in another context. And genes associated with one trait proved to be implicated in a quite different trait too.
• The genome does not control the cell. Rather, it supplies resources for the cell as an autonomous and integrated entity. Genes are not a blueprint. They impart capabilities; the rest is up to us, in interaction with our environment.
• Most human traits are not simply genetic - they are also affected by the person's environment.
• How we are is correlated with our genotype, but genes are not what make us what we are.
• The many genes linked to IQ are sure to be implicated in other traits too - perhaps neuroticism or schizophrenia. There are no isolable intelligence genes. If we select embryos for highly polygenic traits like this, we will have little idea for what else we might be selecting.
• Many human traits are influenced by many genes. Even for traits that show a strong heritability, such as height, the genetic component derives from the tiny effects of many genes rather than big effects from just a few. This can make it very difficult to figure out what the respective genes are doing - or indeed, how causally relevant they really are.
• Sometimes the polygenic nature of traits is extreme, and hundreds or even thousands of genes might be implicated. The statistical associations observed between many complex traits and genes tend to be spread across most of the genome. 62% of common single-nucleotide polymorphisms (SNPs - where one base pair is different) are associated with height in parts of the chromosomes that are active across most cell types, and often in "non-coding" sequences.
• You can't compute from the genome how an organism will turn out, not even in principle. There is plenty that happens during development that is not hardwired by genes. And from a single protein-coding gene, you can't even tell in general what the product of its expression will be, let alone what function that product will serve in the cell.
• Evolutionary and developmental biology are seeing two different kinds of explanation - The first considers population-level phenomena, the second focuses on individuals. Organisms acquire their form twice over: by evolution (the history of which is imprinted in genomes), and by development (through the interactions of molecules and cells). Both are under genetic influence, but not in the same way.
• Genes don't compete with each other. Rather, a mutation to a gene that turns it into a variant that enhances survival of the organism carrying it will tend to spread through a population an eclipse the other less successful alleles. So different alleles of the same gene compete with one another, while different genes on the same genome cooperate. The first is a story about evolution, the second about individual development.
• Genes do not produce life, but on the contrary depend on it.

3. RNA and Transcription: Reading the Message

• The process of transcription:
  • First, each gene is read by an enzyme called RNA polymerase that steps along a single unwound template strand of DNA and assembles a corresponding mRNA molecule use the sam principles of base pairing that holds the double helix together. The double helix is unzipped by the enzyme's advance and reunited in its wake.
  • Once an entire gene has been transcribed into an mRNA molecule, the RNA detaches from the DNA and makes its way beyond the cell nucleus to the place where it can be converted (translated) into a protein by the ribosome.
• Genes - or more generally, functional sections of DNA - are not simply instructions for making proteins. Some of them determine what kinds of proteins cells, tissues, and organs can make. Some do other things: RNA-related things that don't directly involve proteins at all.
• Rather than DNA directly producing proteins, an RNA intermediary creates a buffer layer and gives flexibility and versatility to the readout process from the genome. It also allows many protein molecules to be produced rapidly from a single piece of DNA.
• DNA sequences are full of segments that don't feature in the mRNA:
  • Introns - Are edited out before the mRNA is presented to the ribosome for translation.
  • Exons - Are the retained sequences, spliced back together after the introns are removed.
  • Transposons - Are jumping genes that move about in the chromosomes, with about 65% of the human genome capable of exhibiting transposon behavior.
• The genome was meant to be a static blueprint, but it is dynamic and responsive to environmental stresses - more like an organ or an organism in an organism.
• microRNA - Is smaller than normal genes and may regulate regulate up to 60% of our genes as well as targeting other microRNAs (and a single mRNA can be targeted by many mRNAs). mRNAs seem to control embryo development and to control pluripotency is embryonic stem cells
• Now that the human genome is mapped, the ENCODE project is identifying which parts of the entire genome are transcribed in the cells of different tissues, characterizing the human transcriptome. The HGP showed that only 2% of our genome consists of protein-encoding genes, and most of the rest was considered "junk" (non-coding), but now it seems that genes as protein-encoders are maybe only a small part of what is going on with our genome.
• Encode identified around 37.6k long non-coding RNA (ncRNAs) genes, nearly twice as many as genes that encode proteins.
• It seems that, for mammals, the noncoding, regulatory regions of the genome are more important than the coding parts. Bacteria have 90% of the genome for protein coding, while we have 2%.
• Epigenetics (how your behaviors and environment can cause changes that affect the way your genes work) rather blurs the distinction between nature and nurture.
• Childhood trauma can be correlated with methylation in a person's genome, a biological embedding of experience.
• The gametes (eggs and sperm) have mechanisms that protect their genomes from much epigenetic alteration, and strip away more or less all of it anyway as these cells mature into the state that takes part in reproduction. At the stage of gamete maturation called the primordial germ cell, epigenetic marks in the genome are largely erased.
• Rather than waiting for the genome to evolve, it is more effective to evolve mechanisms for producing rapid revisions of how the genes are used when circumstances demand it.
• Life is according to Dennett and Michael Levin, "cognition all the way down".

4. Proteins: Structure and Unstructure

• There are thought to be around 80-400 thousand varieties of protein molecules in human cells, and they are the second key functional components of life after DNA and RNA
• Proteins are typically dubbed the workhorses of the cell. In this sense they are nearly synonymous with enzymes, the molecular catalysts that conduct life's chemistry.
• Proteins belong to the class of molecules called polypeptides, in which amino acids are strung together in a chain via chemical linkages called peptide bonds. In enzymes, these chains fold up into compact blob-like forms with specific shapes, sculpted by evolution, that allow each protein enzyme to do its job.
• Hydrogen bonding and the hydrophobic attraction are generally thought to be the two key forces that dictate how proteins fold. The "native fold" is the protein's folding target, which is the form with the lowest energy possible.
• Google's Alphafold has made predictions about the structures of more than 200m proteins - all those currently know to biology - in species ranging from bacteria to plants and animals.
• Protein molecules are huge and their structures are extremely complicated.
• Some proteins become embedded in cell membranes, others are structural - they don't function as enzymes, but rather as components of the cell's fabric, and som act not to catalyze reactions, but to regulate genes.
• Because we use crystallography to analyze proteins, and many won't form crystals, we only know about the structures of half of them that make up the human proteome. The rest are sometimes called the "dark proteome".
• Many don't have well-defined folded shapes - maybe up to half in the human proteome, and this disorder seems particularly prevalent in many of the most important proteins in the molecular ecology of the cell. Compare this to bacterial proteomes, where only 4% are disordered.
• Disordered proteins can increase the complexity and versatility of our regulatory networks, but at the cost of increased risk of toxic aggregates formed from misfolded proteins. Many proteins found to be associated with diseases are highly enriched in disoredered regions, including especially those implicated in other neuro-degenerative conditions such as Parkinson's and Huntingdon's.
• On average, each human gene encodes about six different proteins. Some genes may generate many variants, sometimes hundreds (though not all of these may be functional).
• From around 20k genes, our cells can make 80-400k different proteins.
• Gene sequences do not typically encode a specific protein with a specific shape. Instead, they are resources for making families of proteins.
• The amino acids in proteins seem to be organized into domains, and these domains can be mixed and match to produce new and different proteins with the capacity to stick together, to communicate with one another, and to differentiate into specialized forms.
• Just as disorder in protein structure creates flexibility for what a protein can do individually, so random associations in the evolvosome might open up fresh possibilities for what proteins can do collectively.
• Scaffold proteins (SPs) have the job of gathering other proteins together in the same space to communicate. They are good at forging new links between signaling pathways, opening up new developmental possibilities. They are not tailor-made, but good at improvising, often by a reshuffling of the binding domains they possess.

5. Networks: The Webs That Make Us

• Biologists began to think about genes as a network. They were speaking to one another, and the task was to figure out who was conversing with whom: to map out not just the genome, but the connectome through which it operates.
• For complex organisms like us, the network of interacting molecules, the interactome - sometimes seemed not just absurdly but impossibly complicated.
• DNA packaged into chromatin looks more like a cupboard stuffed full of documents in a random heap of loose pages. Yet somehow there is precise cross-referencing and coordination amid the chaos.
• With a combinatorial system you have many more options that will do the job, while retaining distinctions of use and meaning. It's a little like the way languages work: there's nearly always more than one combination of words that will convey much the same meaning, but those combinations aren't arbitrary. In many particular contexts, some choices work better - convey a clearer or more emphatic meaning - than others.
• Perhaps the most useful analogies for how cells work are themselves biological, such as olfaction or cognition. Maybe the only way to truly understand life is with reference to itself.
• It's possible that making promiscuous, reconfigurable networks doesn't just convey advantages but perhaps is the only way a complicated system like our cells can work, if it is to be robust against ineluctable randomness and unpredictability in the fine details.
• Cellular systems are very noisy. Molecular encounters in this crowded, jostling environment are very much a matter of chance, and there are also random fluctuations in the number of different proteins that get produced from moment to moment. That's one reason the cell's wiring can't be compared to the complex electronic circuits in a laptop: no two cells are ever in wholly identical states at any given moment, even when they are both ostensibly doing the same job, such as acting as muscle or kidney cells.
• Combinatorial logic has a certain amount of sloppiness that can absorb (and even exploit) such variation.
• It seems likely that metazoans have evolved this evolvability. Eukaryotes have chosen this sloppiness - probably because it allows new regulatory pathways to develop, opening up the potential for variation and evolvability.
• Causal emergence allows noise reduction - independence of the outcome on random fluctuations or chance events at the microscopic level. And, it makes the causes of behavior cryptic at the microscopic scale, hiding it from pathogens that can only latch onto particular molecules.
• Evolution doesn't so much shift all causation to higher levels as spread it among the various levels.
• Causal emergence seems to be a general design principle for life, but it is rarely evident in our own technologies. Machines tend instead to use simple chains of causation: this cog turns that one.
• Perhaps language is the only human technology that resembles how life works in the causal sense, in which meaning and causal power - the ability to induce thought, mood, action - increase as we go up the scale from letters (or phonemes) to words, sentence, paragraphs, and so forth. Zoom in on a text's component characters and you lose all meaning: the characters themselves are not only effectless but have no intrinsic function.
• The key reason causal emergence seems to be so widespread in how life works is, then, that this is how to "engineer" with noisy components. If you are making a machine from chunky, precision-milled cogs, you don't need causal emergence, because the parts can be relied on.
• The more complex multicellular organisms that appear later in evolutionary history tend to assign causal roles to higher levels of organization in their networks. In this way, these organisms can tolerate more noise and indeterminism in the microscales, because those scales aren't the primary determinant of phenotypic outcomes such as body shapes and behaviors.
• The switch to multicellularity seems to involve the appearance not of more primary genetic resources - more or different genes - but of new ways to regulate them.
• Is there, after all, really such an obvious advantage to being multicellular? If so, we don't know what it is. Complex multicellularity has arisen only twice during evolution: in animals and in plants (fungi might be deemed a third group, but they don't have anything like the tissue diversity of the others).
• Rather than there being smooth and gradual changes in the frequencies of CNEEs (conserved nonexonic elements), three distinct eras of change seem to have occurred over the past 650m years:
  • Until about 300m years ago, when mammals split from birds and reptiles, changes in regulation seem to have happened mostly in parts of the genome close to transcription factors and the key developmental genes that they control.
  • Between 300-100m years ago, those changes tailed off, and instead there were changes near genes that code for the protein molecules serving as receptors of signals at the cell surface. What seemed to matter was a shift in the way cells talk to each other
  • Since 100m years ago, as placental mammals developed, the regulatory changes seem to be associated with mechanisms for modifying protein structure after translation, especially for proteins that are associated with signal transduction within cells.
• Evolution might be considered to have successively discovered ways to innovate and generate new organisms by first reshuffling how developmental genes are switched on and off, then how cells communicate, and then how information gets passed around inside cells. In all cases, the action is taking place not at the genetic level but at higher levels of network organization (which nevertheless leave traces in the genomes.
• Aside from the well-acknowledged complication of viruses, all life is cellular, passes on hereditary information in DNA, and uses proteins and a relatively small palette of other molecular types.
• What seems to have changed over the course of evolution is nothing less than the locus of causation. I have called this causal spreading.
• I think we will find that what gave us cognitive abilities no other species possesses - in particular, the ability to develop language, to think abstractly, and to maintain highly nuanced social interactions - is a change at a higher level than the genomic: a change in the emergent properties of the brain. No single gene "made us human".

6. Cells: Decisions, Decisions

• We're not just a homogenous mass of cells.
• Cells commit to fates by noticing and assessing what their neighbors are doing. It's like voting with a show of hands: we might sneak a look at how others are voting before deciding which way to go ourselves. And if we're given new information, we might change our mind. As new cells are produced by division in the early embryo "they take care of their own further development, shaping both themselves and their local environments without any further instruction from their parents.
• Each cell inherits the epigenetically modified genome of its parent cell, in which some genes have been silenced while others are active - but it can incur further epigenetic modification, guided by the signals coming from outside, that change its fate still further. In this way, the development of the organism is a story about progressive cell-fate selection, at each stage of which the cell lineage commits to one path and the others become unavailable to it.