Biology · Gene Regulation & Cell Signalling
Feedback, switching and interacting regulatory networks
A material pool stays near the same amount while molecules continuously enter and leave. After a loss, the incoming flow increases. The pool's amount and its traffic are different records, and the return effect on that traffic forms the feedback loop.
The gene-expression chapter explains transcription, RNA handling and translation; biological energy explains enzymes, ATP coupling and metabolic feedback. The cells chapter supplies membrane and compartment ideas. Those are the useful foundations here. Each lesson reintroduces its essential terms, and the constructed cases need only simple multiplication, addition and fractions.
Back to the Biology mapBy the end, you can
Distinguish stabilising feedback, reinforcing feedback and constructed memory states without equating homeostasis with inactivity.
Negative feedback opposes a deviation
Feedback occurs when a downstream consequence influences an earlier part of a process. In negative feedback, the return effect counteracts a change. A low pool can stimulate input, or a high pool can reduce it; negative does not mean every response must be a decrease.
Homeostasis is maintenance of relevant internal conditions within workable ranges through continuing adjustments. It does not require one perfectly constant value or a stopped system. Conditions can fluctuate, and the range or regulated setting can change with context. A steady amount can coexist with substantial molecular turnover.
A cell can implement feedback through enzyme regulation, transport, signalling or gene expression. There need not be a little conscious decision-maker or a brain inside the cell. The diagram's sensor/regulator roles name molecular tasks, not intentions.
A regulatory return can adjust ongoing material traffic
Read the diagram in words
A material-flow arrow enters a pool and another leaves it. A dashed sensing link runs from pool amount to a regulator, and a return activation arrow runs from the low-pool regulator to the incoming-flow process. A disturbance arrow removes material. Counts beside separate interval panels show 100, 90 and 98 without forecasting a final equilibrium.
| Component |
|---|
| Incoming-flow process |
| Selected material pool |
| Outgoing-flow process |
| Low-pool sensing/regulatory machinery |
| Specified extra loss |
| From | To | Connection meaning | In this model |
|---|---|---|---|
| Incoming-flow process | Selected material pool | Material flow | Material enters |
| Selected material pool | Outgoing-flow process | Material flow | Material leaves |
| Selected material pool | Low-pool sensing/regulatory machinery | Sensing | Pool level influences regulator |
| Low-pool sensing/regulatory machinery | Incoming-flow process | Activation | Low pool increases entry |
| Selected material pool | Specified extra loss | Material flow | Extra material removed |
| Specified interval | Initial amount | Amount entering | Amount removed | Ending amount |
|---|---|---|---|---|
| Ordinary interval | 100 | 12 | 12 | 100 |
| Extra loss | 100 | 0 | 10 | 90 |
| Specified response interval | 90 | 20 | 12 | 98 |
- Pool units
- Illustrative reference amount: 100
- Stable amount can have continuing traffic
- Return effect opposes the loss
- No complete controller or final steady-state forecast
Positive feedback reinforces a state change
Positive feedback strengthens a change rather than counteracting it. For example, an active regulatory product can help promote further production or activation in its own pathway. Reinforcement can support a switch-like response, but a positive arrow alone does not prove two stable states or permanent memory.
Bistability means that a system can settle into either of two stable states under the same fixed external conditions, depending on history. Our discrete activity rule provides a transparent invented example. Each step keeps half the present activity and adds half a target value: 0 when activity is below 5 units, or 10 when it is at least 5.
Starting at 2 gives 1, then 0.5, then 0.25, tending towards 0. Starting at 8 gives 9, then 9.5, then 9.75, tending towards 10. Both tracks follow the same rule. A sufficiently strong perturbation can move the system across the threshold; persistence is therefore conditional, not immunity to all future inputs.
One specified rule can support two stable limiting states
Read the diagram in words
An activity node feeds a threshold-dependent reinforcing process and also has a loss/relaxation contribution. A separate step plot shows two trajectories under the same external conditions: 2, 1, 0.5, 0.25 towards zero and 8, 9, 9.5, 9.75 towards ten. A horizontal guide at five marks the rule's threshold.
- Specified update rule
- Next activity = half the current value + half the selected target; target 0 below 5 and target 10 at or above 5
- Rule threshold
- 5
| Component |
|---|
| Current activity |
| Feedback-selected target |
| Partial retention/relaxation rule |
| From | To | Connection meaning | In this model |
|---|---|---|---|
| Current activity | Feedback-selected target | Regulatory effect | Threshold selects target |
| Feedback-selected target | Current activity | Activation | Reinforces the high-state branch |
| Partial retention/relaxation rule | Current activity | Regulatory effect | Only half current activity retained per step |
| Model step | Start below threshold | Start above threshold |
|---|---|---|
| 0 | 2 | 8 |
| 1 | 1 | 9 |
| 2 | 0.5 | 9.5 |
| 3 | 0.25 | 9.75 |
| Quantity | Displayed axis minimum | Displayed axis maximum |
|---|---|---|
| Activity units | 0 | 10 |
| Series | Line style | Marker |
|---|---|---|
| Start below threshold | Solid line | Circle |
| Start above threshold | Dashed line | Square |
- Model step
- Activity units
- Rule threshold: 5 units
- Same fixed external conditions
- Stable limits in this model: 0 and 10
- Activity units are not conserved molecule counts
The whole loop matters
Feedback systems can overshoot, fluctuate or oscillate when timing, delays and interacting processes matter. Negative feedback is often stabilising, but its sign alone does not prove stable behaviour. Positive feedback can be bounded by saturation, removal or other loops; it need not grow without limit.
A network may share regulators across several outputs. Altering one component can therefore affect neighbouring routes, and a labelled pathway is usually only a selected view of a larger system. To explain a response, identify the measured variable, the relevant loop and the timescale before making a stability or causality claim.
Worked example
Separate stable amount from changing flow
The ordinary interval begins with 100 pool units and has 12 entering and 12 leaving: it ends at 100. An extra loss of 10 then leaves 90. In the next specified interval, the low-pool condition raises entry to 20 while 12 leave, giving 90 + 20 − 12 = 98. The deviation from 100 shrinks from 10 to 2 units.
This is a constructed example of a response opposing loss, not a full prediction that the next interval must reach exactly 100. For the separate activity switch, compute the high branch's first step as half of 8 plus half of 10: 4 + 5 = 9. Activity units are not molecule counts or a conserved material pool, so the two models answer different questions.
Check your understanding
Can increasing an incoming flow be part of negative feedback?
Show reasoning
Yes. If the pool has fallen, increased entry counteracts that deviation; negative refers to the return effect on the change.
Does a pool remaining at 100 units mean no molecules enter or leave?
Show reasoning
No. Equal incoming and outgoing amounts can maintain the pool while turnover continues.
Does drawing a positive-feedback arrow prove that a natural system is permanently bistable?
Show reasoning
No. Stable alternatives depend on the full dynamics and conditions. The supplied switch rule proves only its constructed model behaviour.
What to carry forward
Evaluate feedback through the complete return effect, timing and conditions, rather than through an arrow sign alone.
References
Resources supporting this lesson’s explanations and examples.
- Biology 2e — Homeostasis (2018)Mary Ann Clark, Matthew Douglas and Jung Choi · OpenStaxOpen textbook chapter
- Self-perpetuating states in signal transduction: positive feedback, double-negative feedback and bistability (Current Opinion in Cell Biology, 2002; 14:140–148)James E. Ferrell JrResearch abstract
- Encoding and decoding cellular information through signaling dynamics (Cell, 2013; 152:945–956)Jeremy E. Purvis and Galit LahavResearch review — indexed passages inspected