Monday, May 27, 2024

Thoughts on a podcast about How Life Works


The podcast run by the computer company Oxide has hosted a discussion of my book How life Works. It’s deeply thoughtful, engaged, and also fun, and I’m very grateful for it. I’d recommend it to anyone as a fascinating listen for its own sake.


I’m also particularly grateful to the biologist on the panel, Greg Cost, for such a considered appraisal. Greg doesn’t agree with everything in the book (and he’s a Dawkins fan, after all!), but he is generous and measured in all that he says. 


I have some comments on some of that. One small issue: while it is of course true that cancer is mostly linked to genetic “breakdowns” of some sort, this is not invariably so.


What strikes me most about Greg’s comments – given that he is someone who is evidently very knowledgeable, thoughtful, and receptive – is his pessimism about figuring life out, at least until we have pinned down every last detail. I was pleased and relieved to hear that he recognizes the significance of the various new aspects that I cover, such as the importance of RNA, protein disorder, condensates etc. But it seems that for him this just makes the matter all the more complicated – even though I suggest in the book ways in which we can start to integrate these considerations into a synoptic view that does after all seem to have general principles (just not the ones we once believed!).


It's not that Greg seems to think the general considerations I identify are wrong or misplaced. He simply doesn’t remark on them at all, as if they were not there. I’m struck by this because it’s rather similar to one or two other reviews of the book I’ve had from thoughtful and receptive biologists: they say yes, yes to all the details, but are simply silent about – as if blind to – efforts to draw broader insights from them.


Thence Greg’s view that made what we’ll need is Sydney Brenner’s approach of exhaustive simulations down to the last molecule. That, to me, is the strategy of desperation: “We’ve run out of ideas so we just have to be literal about it and include everything”. What I want the book to suggest is that, not only is this not necessary but it is not helpful. It gives no real insight, but also it denies the fact that no highly complex system can work if every detail matters. What the “new biology” is telling us is precisely that not all the details do or can matter – and what principles might be needed that allow a system to transcend its microscopic details.


Greg, for example, says that his heart sank when we started to realise that the formation of condensates/phase-separated blobs by proteins and RNAs is a common process in our cells. It seems that to him this merely drives another nail into the coffin of the old picture of all things happening in aqueous solution. But I have been immensely excited and stimulated by it. As I suggest in the book, what we’re surely seeing here is a general principle whereby cells create structure and marshal many molecules into a shared location where they can work collectively. Perhaps it’s simply that this is a “condensed matter” way of thinking that is (as I have discovered) wholly unfamiliar to biologists. But I can’t help but suspect too that all biologists tend to see is a further receding of long-cherished ideas. I was struck by Greg’s apparent longing to retain the “DNA blueprint” picture, seemingly by means of saying that there is after all still a blueprint but also lots of “noise” between it and the phenotype. To my mind, that “noise” is metaphorical, perhaps even psychological: it implies “details we don’t and may never understand, but which can be invoked to explain away anything that doesn’t fit a blueprint picture”. The story in the podcast about the building of household steps was telling: in the end it amounted to “Well, the steps were made according to a blueprint, even though the blueprint was ignored”. The actual construction, the application of a real joiner’s skill in seeing how to do the job, then became the “noise” that accounted for the difference between the alleged blueprint and the reality. In other words, “noise” here is the gap between how we see life work, and how we would like it to work.


This attitude becomes particularly evident in Greg’s comment about theoretical biologist David Penny’s remark from the book that he’d have been proud to be on the committee that designed the E. coli genome, but not the human genome. Greg takes this to imply that the genomes of bacteria/prokaryotes are somehow “more evolved” and thus better designed, as though the 3.5(?) bn year evolutionary history of prokaryotes has perfected their genomes.


I’m very curious about this, because it is so obviously wrong (as well as missing the point) that, coming from a very smart guy, it seems to suggest that something else is going on.


The implication seems to be that prokaryotic genomes were once as seemingly messy and confusing as ours, but have been cleaned up and streamlined by their longer evolution. But that’s simply not so. After all, no genome was produced de novo since life began. Rather, we obviously evolved from prokaryotes, and so our genomes have acquired their alleged messiness out of the simplicity of theirs. Indeed, molecular phylogenetics allows us to effectively watch this happen. 


So the puzzling complexity and seeming messiness of our genome is absolutely not something that, given time, evolution could and would simplify. Rather, this difference seems to have been *necessary* to make complex metazoans possible.


So it rather feels as though Greg is desperately eager to find a way of making biology “make sense” – which E. coli seems to promise with its more transparent logic. This then becomes how biology “should be”, while we are a messy aberration. That is reinformed by the fact that, when one of the panellists raises my point that it is evidently now known that what is true for E. coli is not necessarily true for the elephant, Greg immediately and only talks about the ways in which we are the same.


As I say in the book, Penny’s remark reveals how far the answer of “how life works” for us humans is from our intuitions based on machine-like technologies. The fact that we look at it and think “Good God, what?!” is not an indication that nature has done a terrible job, but that we need to rethink our preconceptions. And that’s what this book is about.


Oh, and guys, guys: yes, you’re computer scientists so you know all about Alan Turing. But I’m not sure you know how much he was (and perhaps still is) not known beyond your field. And yes, as history The Imitation Game was pretty terrible - but that doesn’t mean it didn’t help to put Turing on the radar of lots more people. Just as it may be inconceivable to you that plenty of people don’t know about Turing, so it is not easy for me to accept that some people who know a lot about Turing did not know at all about his morphogenesis paper. We’re all constrained by our priors!


But in any event, thank you chaps for a nice discussion.