Skip to main content

Biological News

Reductionism Pays My Mortgage

Reductionism has been a powerful tool for a long time now, and it’s not going away any time soon. There, that should be enough of a 50,000-foot opening! The idea that you can understand a system by taking it apart and looking at the pieces doesn’t always work out, but it still seems to be the best way to go about it. And even if said system turns out to be more than just the sum of its parts, knowing what those parts are and what their immediate functions are is not a step that can be skipped, either.

Consider (philosophy again!) that all we know about the physical world is from what we can take in from our senses. How, exactly, do we do that? We’ve known for a long time now the main details of how light signals get sent to the brain: the eyes have a light-sensitive layer, the retina. When looked at more closely, the retina turned out to have particular cells, rods and cones, that were responsible for black-and-white and color vision (as an amateur astronomer, I’m used to extended sessions where my cone cells do hardly any work at all, since most deep-sky objects are too dim to set off any color vision response). When looked at more closely in turn, these cells turned out to contain particular pigments that were somehow light-sensitive, and when those were looked at still more closely, everything was found to hinge on photochemical isomerization of carbon-carbon double bonds. Reductionism in action – you start by wondering about your eyeballs, and you end up talking about molecular orbitals.

Similarly, sound is transduced by the actions of the small bones of the inner ear, and then by the hair cells therein. But past these two senses, things get a bit less obvious. It’s clear that through our skin we can sense relative temperatures, along with pressure and other kinds of touch but how, exactly, do we do that? Eventually, these things are also going to have to come down to molecular-level mechanisms, just like vision. There are, in fact, ion channels that are sensitive to physical deformations of the cell membrane, several classes of them, and these seem to be largely responsible for the molecular-level sensing of touch, pressure, and so on. (They’re also involved in hearing, when you get down to the level of asking how the movement of the individual hairs in the hair cells gets turned into some sort of cellular signal). These same types of receptor, and the specialized cells bearing them, also show up in things that we don’t sense directly at all, such as the regulation of blood pressure.

Temperature sensing seems to divide out into separate classes of hot- and cold-sensing receptors, but it appears that these may work through a common mechanism. That link is another trip down the reductionist manhole – you start off wondering how you can sense that something is hot or cold, and you end up thinking about entropic/enthalpic balancing in protein conformational states and changes in molar heat capacity, which is what it all comes down to. As you might figure, it turns out that compounds like menthol activate a particular cold-sensing channel, while capsaicin from peppers activates a heat-sensing channel, so the English language’s confusion between “temperature hot” and “spicy hot” is an honest one.

Since I work in the drug industry, reductionism is the reason I still have a job. The idea, drained of all human feeling, is that individual drug substances can make enough of a difference in the health of a person so that said person will be motivated enough to pay money for them. The reductionist pathway is person/organ system/cells/biochemical pathway(s)/individual proteins/small molecule ligands, and it works well enough, at times, to pay the bills and save some lives. Long may it continue to!


17 comments on “Reductionism Pays My Mortgage”

  1. Anon says:

    “The idea that you can understand a system by taking it apart and looking at the pieces doesn’t always work out, but it still seems to be the best way to go about it.”

    That’s fair enough, but what if our ultimate goal is just to develop drugs, rather than understand everything? Perhaps trying to unravel the mysteries of the universe is just a distraction from doing our job?

  2. I would say that reductionism is also why our jobs are getting harder and why we are in trouble. It’s why we are discovering the limitations of target-based drug discovery and why phenotypic screening has become the rage. As usual, too much of a good thing can be a bad thing.

  3. Cellbio says:

    While I generally agree, could say the mortgage payment is assured through empiricism more than reductionism. The reductionist approach yields the molecule and rationale for the bet, empirical science generates the risk/benefit ratio that yields the rationale for payment.

    The other issue, common to approaches that yield success, even if black swans, is the approach is over valued imo. here I don’t think the issue arises when walking down the reductionist path, but lies in the lack of appreciation for the nature of the empirical nature of the clinical bet, namely that one examines whether the chosen molecule is safe and efficacious, which to varying degrees test the reductionist hypothesis. In some cases, the pharmacology of the drug candidate may be so tight the hypothesis is tested well, but in others, we presume to much and narrow or thinking due to reductionism when there is a lot we do no know about the pharmacology.

    Did pravastatin reduce CV events in the West of Scotland study because of cholesterol lowering or because of the total pharmacological impact that includes unappreciated pharmacology? Gleevac is a result, I believe, of really good reductionist thinking, but efficacious because of more than abl activity. These examples do more to support reductionist thinking than refute, but offer a fair premise that empirical testing of a pharmacological agent is really where the money is made. For instant, last I knew, it has been a while, anti-CD20 antibodies that cleaned up the chimeric antigenicity resulted in less efficacy, suggesting a host immune response to the chimeric antibody contributes to efficacy.

    And, what about the reductionist approaches that yield great rationale for bets that fail? Are these not the norm rather than the rule? One should ask whether the reductionist approach is working at scale. Is it viable but over-played? I think some mortgages might be in jeopardy.

    Program and company failure rate have led me to conclude that earlier and more thorough empirical testing to compliment the clean reductionist theories should be widely deployed. When i had a chance to do this, it was eye opening to see that a few thousand inhibitors of an enzyme, when tested broadly in biological systems, had many many many pharmacological profiles. We know a lot less about a drug candidate than reductionism leads many people to believe.

  4. Hap says:

    It’s hard to know when the behavior of a system is not predictable from the behavior of its parts without knowing how its parts behave (and what they are).

  5. anonymous says:

    If you were fully reductionist you could design a perfect drug on a computer

  6. CS says:

    The reduction of biology to chemistry certainly works some of the time, but it’s a major mistake to think that it’s a general answer. An organism works at its own level, and its chemistry is determined as much by macroscopic factors as microscopic ones (think adrenaline at the sight of danger, changes in brain chemistry after an emotional trauma, etc. etc.). Macroscopic features had to be invoked in all the examples above, and in any case the actual use of those senses isn’t reducible – the point of sight is the information in contains, and our physical structure has developed driven by that, rather than by chemistry.

  7. Peter Kenny says:

    Always a giggle to ask reductionists (and systems biology types) about free intracellular concentration…

  8. biotechtoreador says:

    “reductionism is the reason I still have a job”

    Well, reductionism and biopharma’s ability to extract massively high prices for its products….

  9. Anon says:

    One word: Mindwank = intellectual self-stimulation.

    Personally I prefer just to discover drugs, and I’m as happy for them to come from luck as from design, because my ego is less important than paying my employer’s bills. Once I pay those, I know they are happy to keep paying my mortgage…

    By the way, how many drugs has Vertex discovered?

  10. Hap says:

    I think your schtick has already been responded to…. (you can substitute “patents” for “journal articles”, and, no, disclosures on don’t count).

  11. steve says:

    I thought “reductionism” referred to the dwindling number of jobs available.. 😉 Anyway, I got my concept of reductionism from one of my fellow postdocs who was a physicist – he said that Biology is just Chemistry, Chemistry is just Physics, Physics is just Math and Math is God. I told him that the concept of God is a mental construct from humans so God is just Biology. (I also tried to explain to him that he shouldn’t be so arrogant about physics and math when they can’t even solve the three-body problem….).

  12. ScreeningAgnostic says:

    Right on Derek. Thank you for trying to reality-check.

    Ash – “phenotypic screening has become the rage” – is that a recommendation, or just noting another fad? The next time someone cites Swinney and Anthony to me I swear i’ll.. (Sorry David)

  13. Ash says:

    ScreeningAgnostic – Just noting another fad. As with any other tools rumors of its efficacy are grossly exaggerated, but it’s still a good tool to have in your armamentarium of arbitrary accessories.

  14. Nomen Nescio says:

    Two objects may have the exact same temperature, yet one may feel cold and the other, say, room temperature warm. What we really sense is the flow of heat.

  15. cthulhu says:

    Nobel laureate Robert Laughlin (physics, the fractional quantum Hall effect) wrote a fascinating book about the limits of reductionism that gets into the interfaces between physics, chemistry, biology, etc. Here’s a link:

  16. Julien says:

    Phenotypic screening is also a product of reductionism.

Comments are closed.