Necessary and sufficient are elements essential for constructing causality in biology.
One way to contemplate necessary and sufficient in biology would be to consider whether suspected cause always precedes outcome (is it necessary) and whether suspected cause always produces outcome (is it sufficient).
Brief History Of Causality In Biomedicine
How to separate chance from causation in biology? Consider aspirin and pain. Have pain, take aspirin, pain disappears, i.e., clearly sequential, a key precept of causality (). Though swallowing aspirin doesn’t always yield pain relief, it has done so enough times in enough people for their association to be not by chance or coincidence.
Causality entered the realm of biomedicine with the Henle-Koch postulates (), commonly attributed to in 1840 and Robert Koch, his student, in 1890 (3). These postulates were instrumental in illuminating the causal relationship between a particular microbe and a specific disease, and in bringing scientific rigor to medicine through the notion of causality. Though these postulates haven’t withstood the test of time unchanged, having undergone numerous revisions and becoming studded with umpteen caveats ( , , 6, 7), nevertheless they epitomize a revolutionary scientific breakthrough that eventually transformed medicine. As Robert Sackstein states ( ),
‘Koch’s postulates infused scientific rigor into medicine, altering the cultural foundation of medical science from that of observation/association/correlation towards one grounded in causal relationships yielding mechanistic insights’
A recent salient example is that of the 2005 Physiology/Medicine Nobel Prize winnerportrayed in the unabashedly hagiographic dramatization below ( ) as a sort of ultimate truth-seeker who didn’t hesitate to infect himself to prove caused peptic ulcer.
Today this notion of causality permeates the entire biomedical research landscape, bestowing primacy to questions of mechanism of action. Even inwhere the notion of causality typically remains foolhardy at the best of the times, postulates such as the , originally formulated by , attempt to decipher a causal association from among multiple factors and a particular biological outcome, typically disease ( ).
Problems With Biological Causality Or Why Necessary And Sufficient Came To Be
In its simplest form, causality can be stated as A causes B. For example, step under a waterfall (A), get wet (B). However, such relatively simplistic, linear relationships are usually outliers and not the norm in biology.
- Imagine a scenario where a particular bacterium must acquire a through horizontal gene transfer of discrete gene segments to become virulent (necessary) to a particular animal and doing so should always yield (sufficient). Sufficiency is thus a minimum requirement to recapitulate a biological feature, in this case virulence.
- Acquisition of a pathogenicity island in this particular case could be both necessary and sufficient to bestow virulence if experiments showed inactivation of genes within it caused a measurable loss of virulence ( ).
Causality becomes even trickier to decipher in epidemiology.
- For example, a relationship between smoking (A) and lung cancer (B) is now well-established. However, to say smoking causes lung cancer isn’t strictly accurate because not every smoker develops lung cancer, which by the way can also occur in those who never smoked. The reality is smoking isn’t sufficient to induce lung cancer but can increase the probability of developing it. Thus, smoking could be somewhat necessary but not sufficient to cause lung cancer, simply because it’s more common among smokers than non-smokers.
- Another cause and effect example from cancer is the case of HPV ( ) (A) and (B), where the association is much stronger though still not absolute, since not all HPV infections result in cervical cancer. The WHO states ( )
‘Nearly all cases of cervical cancer can be attributable to HPV infection’ .
Such examples thus help break down biological causality into two distinct components,
- Does HPV (A) always precede cervical cancer (B), i.e., issue of necessity? Here the answer is a qualified yes.
- Does HPV (A) always produce cervical cancer (B), i.e., issue of sufficiency? Here the answer is no.
In a recent example, by painstakingly knocking out ~5400 genes in yeast two at a time, creating ~23 million yeast strains over a 17-year odyssey, biologists at University of Toronto identified those genes that when removed as pairs, resulted in sickness or death (, see below from ), i.e., both necessary and sufficient for viability. ~1000 yeast genes are already known to be essential for viability individually. This process uncovered ~3300 additional genes.
By assessing the relative contribution of each factor involved in a particular biological process and thereby helping to more accurately pinpoint the nature of their involvement, necessary and sufficient become the critical means for attempting to establish biological causality. This is because unlike the examples that made 19th century Bacteriology its Golden Age, most biological phenomena and diseases cannot be reduced to a simple, linear cause and effect between one particular agent (cause, A) and biological outcome (effect, B). Rather, most entail the interplay of multiple, overlapping, maybe even redundant, not necessarily contemporaneous factors where many individual ones may be necessary but not sufficient.
Causality’s Value In Biomedicine
The English philosopherstated (15),
‘the cause of an event in nature is the handle so to speak, by which we can manipulate it’
The hope is the greater details with which biological factors can be parsed as necessary and/or sufficient, greater the capacity to better understand where, when, how to act on, intervene in and/or manipulate what () to better prevent, treat or cure illnesses.
1. Hara, Kenta, et al. “Amino acid sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common effector mechanism.” Journal of Biological Chemistry 273.23 (1998): 14484-14494.
2. Morabia, Alfredo. “Hume, Mill, Hill, and the sui generis epidemiologic approach to causal inference.” American journal of epidemiology 178.10 (2013): 1526-1532.
3. Evans, Alfred S. “Causation and disease: A chronological journey The Thomas Parran Lecture.” American Journal of Epidemiology 108.4 (1978): 249-258.
4. Fredericks, D. N., and David A. Relman. “Sequence-based identification of microbial pathogens: a reconsideration of Koch’s postulates.” Clinical microbiology reviews 9.1 (1996): 18-33.
5. Falkow, Stanley. “Molecular Koch’s postulates applied to bacterial pathogenicity—a personal recollection 15 years later.” Nature Reviews Microbiology 2.1 (2004): 67-72.
6. Gradmann, Christoph. “A spirit of scientific rigour: Koch’s postulates in twentieth-century medicine.” Microbes and Infection 16.11 (2014): 885-892.
7. Byrd, Allyson L., and Julia A. Segre. “Adapting Koch’s postulates.” Science 351.6270 (2016): 224-226.
8. Sackstein, Robert. “Fulfilling Koch’s postulates in glycoscience: HCELL, GPS and translational glycobiology.” Glycobiology (2016): cww026.
9. Marshall, Barry. “Helicobacter connections.” ChemMedChem 1.8 (2006): 783-802.
10. Hill, Austin Bradford. “The environment and disease: association or causation?.” Proceedings of the Royal society of Medicine 58.5 (1965): 295.
11. Falkow, Stanley. “Molecular Koch’s postulates applied to microbial pathogenicity.” Review of Infectious Diseases 10.Supplement 2 (1988): S274-S276.
13. Costanzo, Michael, et al. “A global genetic interaction network maps a wiring diagram of cellular function.” Science 353.6306 (2016): aaf1420.
15. Collingwood, R. G. “Causation in practical natural science.” RG. Collingwood, An Essay on Metaphysics. Revised Edition. R Martin, ed (1940): 286-312.
16. Gillies, Donald Angus. “Establishing Causality in Medicine and Koch’s Postulates.” (2016).