19.2 Where Does Biological Randomness Come From?

19.2 Where Does Biological Randomness Come From?

The previous section showed that genetically identical cells can exhibit remarkably different behaviors. Before developing mathematical models of this variability, we first need to understand its biological origin.

A useful first step is to distinguish between two fundamentally different sources of randomness.

Some variability originates outside the biological system.

Other variability is generated inside the system itself.

This distinction plays a central role in stochastic modelling because the two sources of randomness often require different mathematical descriptions.

Extrinsic randomness

Biological systems rarely exist in perfectly controlled environments.

Even under carefully designed laboratory conditions, cells experience small differences in their surroundings.

Examples include

• slight variations in nutrient availability,
• local temperature fluctuations,
• changing oxygen concentrations,
• signaling molecules released by neighboring cells,
• or differences in mechanical forces.

These factors influence the behavior of the cell without being generated by the cell itself.

Such variability is called extrinsic noise because its origin lies outside the regulatory system under investigation.

From a modelling perspective, extrinsic noise can often be viewed as uncertainty in the environment or in the values of model parameters.

Intrinsic randomness

Even if every cell experienced exactly the same environment, variability would not disappear completely.

Many biological processes are inherently stochastic because they involve the interactions of individual molecules.

Consider the transcription of a gene.

A transcription factor must first bind to a promoter before RNA polymerase can initiate transcription.

Whether this binding occurs during the next millisecond is fundamentally a probabilistic event.

The same is true for

• transcription,
• translation,
• protein degradation,
• molecular diffusion,
• and many biochemical reactions involving only a few molecules.

When only a handful of molecules participate in a reaction, random fluctuations become unavoidable.

This variability is known as intrinsic noise because it originates from the molecular dynamics of the biological system itself.

Why molecule numbers matter

The importance of intrinsic noise depends strongly on the number of molecules involved.

Imagine flipping a single coin.

The outcome is completely uncertain.

Now imagine flipping one million coins.

Although each flip remains random, the fraction of heads will be very close to fifty percent.

The random fluctuations average out.

The same principle applies to biological systems.

When millions of molecules participate in a reaction, deterministic models usually provide an excellent approximation.

When only a few molecules are present, however, random fluctuations become comparable to the average behavior.

Under these conditions, stochastic models become essential.

Noise is not always undesirable

The word noise often suggests an unwanted disturbance that should be eliminated.

In biology, however, stochasticity can be beneficial.

Random fluctuations generate diversity within genetically identical populations.

This diversity allows populations to explore multiple physiological states simultaneously.

For example, a small fraction of bacterial cells may randomly enter a dormant state, making them more likely to survive antibiotic treatment.

Similarly, stochastic gene expression can influence developmental decisions, cell differentiation, and stress responses.

Rather than representing a failure of biological regulation, randomness often contributes directly to biological function.

From variability to probability

Understanding that biological variability has both intrinsic and extrinsic origins changes the way we think about modelling.

If randomness is an essential component of the biological mechanism itself, predicting a single trajectory is no longer sufficient.

Instead, we must ask questions such as:

• How likely is a particular outcome?
• How variable are repeated experiments?
• Which probability distribution best describes the observations?

These questions require a new mathematical language based on probability rather than deterministic trajectories.

Before introducing this language, however, we first consider a remarkable historical experiment that fundamentally changed our understanding of biological randomness.

The experiment demonstrated that probability distributions themselves can distinguish between competing biological mechanisms.

Key concepts

• Biological variability originates from both intrinsic and extrinsic sources.
• Extrinsic noise arises from fluctuations in the environment.
• Intrinsic noise originates from stochastic molecular events.
• Random fluctuations become increasingly important when only a few molecules are involved.
• Biological noise can have important functional consequences.
• Understanding variability requires thinking in terms of probability distributions.

Summary

Biological randomness has multiple origins. Extrinsic noise reflects fluctuations in the cellular environment, whereas intrinsic noise arises from the stochastic behavior of individual molecules. The relative importance of these two sources depends strongly on the scale of the biological system. At low molecule numbers, random fluctuations become a fundamental aspect of biological regulation rather than a negligible perturbation. Recognizing this distinction motivates the development of stochastic models that predict probability distributions instead of single trajectories.

Self-check questions

1. What is the difference between intrinsic and extrinsic noise?
2. Why does intrinsic noise become more important at low molecule numbers?
3. Give three examples of extrinsic noise.
4. Give three examples of intrinsic stochastic processes.
5. Why can biological noise be advantageous for a population?
6. Why does understanding biological variability require probability distributions rather than single trajectories?