top of page

Modern Biology: Embracing Uncertainty

Updated: Mar 25

Sabrina Fernando shares a quick overview of the uncertainties in biology: the oftentimes lack of concrete or absolute truth; and gives a few common examples of how to reckon with it.

Introduction to Uncertainty through Modern Physics

The year was 1905, and a 26-year-old clerk in Switzerland published four papers. The clerk was none other than Albert Einstein. And the four papers -later known as the ‘miracle year papers' - are the very papers that laid the foundation for modern physics. This, and several other discoveries in the twentieth century regarding the quantum century and theories of relativity transformed how we describe physical science today. This shift in perspective became highly useful on “smaller scales.” For example, as the scientists peered deeper and deeper inside the atom: classical mechanisms, too precise and too deterministic in the equations of motion and mass failed; where quantum physics, content with measuring probabilities succeeded.

“Modern” Biology: More Molecular, More Mathematical, More Uncertainty

Biology experienced a similar shift in the later part of the twentieth century. The modern era of biology started with the discovery of the double-helical structure of DNA in the 1950s. The eventual sequencing of the complete human genome in the early 2000s and the development and mainstreaming of one molecular technique after another(cloning! rtPCR!! Next-gen sequencing!!! rnaSEQ!!!!) quickly saw modern biologists traversing in seas of data. Today, biology is molecular, data-driven, and thus, reliant on probabilities, models, and numbers. Today’s biologists no longer report a singular number without an expression of uncertainty accompanying it. What contributes to this addition of uncertainty to biological measurements?

A Practical Example of a Common but Complex Disease

In the 1930s when it was proved that a patient can become diabetic either from being deficient in insulin or becoming resistant to insulin, it was done through an experiment (of fasting the samples). However, the molecular factors that predispose someone to develop either form of diabetes were not clear. It is now, after the post-genomics era, that we can compare the variants of certain locations in human genomes between control and patient samples and describe which variant(s) is more likely to be present in diabetic patients. In the cases of complex diseases such as this (and much of modern biology) nothing is simplistic. The likelihood of a person developing diabetes is a function of not just variation in a singular location in the genome, but multiple locations, how they interact with each other, and how they interact with the environment. Also, as this sort of molecular measurement looks at probabilities, there will be an expression of uncertainty attached to it. We cannot say a person having X risk variant will develop Y condition for certainty. We can only say, having X risk variant gives one A% chance of developing Y condition, and we have B% confidence in our estimate.

Modern Biologists: Express Uncertainty with Clarity!

How do we reckon with modern biology and its uncertainties? First and foremost, young (and aspiring) scientists should learn to embrace it and communicate it in their research output. Going back to the example with diabetes, if you were to compare the body weight of the control versus type 2 diabetic patients in a sample, do not simply report the average/means of each. The mean of the sample you are reporting does not represent the “true mean” of the population. Therefore, it is more appropriate to attach a 95% confidence interval to it, which shows a range of values around the sample mean that you can be 95% certain of the population mean. If you wish to describe the expression level of a certain gene in control versus patient, again, do not simply report the median value of the expression. Show the distribution of the expression levels in box plots. If you conclude that the body weight of diabetic people differs from that of the control group, describe how significant that difference may be. A smaller P-value, such as p<0.001 tells us that the difference you observed between the two groups can be observed by chance only once if you do the same analysis 1000 times, which makes your report significant.

Modern Consumers: Careful in Our Thinking

As consumers and readers, it is important to remain rational in our thinking. If we ever see claims like X drug has cured cancer or Y receptor holds the secret to treating a disease, we should always remember, modern biology is more complex than that. Therefore, we should not trust clickbait claims, should try to locate the primary literature, and must always look for the numbers.


While not everyone is going to be a scientist, everyone should choose to think scientifically. Biological science is an inseparable part of our lives now. As we read the labels on the food we buy or browse over the acne treatment that seemingly solves everything, we will hopefully look at the range of numbers instead of the claims. And those who will venture into the world of biology, hopefully, will embrace the uncertainty and express it in their research output with grace and clarity.

About Sabrina

Sabrina Fernando fell in love with molecular biology as a teenager. After her undergraduate training in Bangladesh, she came to Australia for her graduate studies. Her research involved exploring the biology of serotonin receptors. In recent years, she has developed an interest in education on mental health and responsible science communication.

All views/opinions shared in this article are the views/opinions of the writer.

bottom of page