eduKateSG · Genetics · The Gold Standard of Education
Decide whether a clone grew by selection, drift or the way it was sampled
Choose the explanation you need, or follow all twelve chapters in order.
Full chapter index · Independent practice · Genetics subject guide
Clonal expansion is an observation: descendants carrying the same inherited marker occupy more of a sample than other marked descendants. Selection is one possible cause, but neutral birth-and-loss, tissue geometry, founder timing and recovery bias can create the same picture. The gold-standard habit is therefore to reconstruct the observation before naming its cause.
The examples below are invented. They connect clone counts to explicit compartments, time points and sampling frames, then ask which mechanisms remain compatible. A large clone is not automatically dangerous, adaptive or even large in the whole tissue. It is large relative to a stated denominator.
This article continues the lineage-tracing route by moving from ancestry to population dynamics. It treats clone size as data generated by a biological process and a measurement process together.
Continue from The Gold Standard of Genetics within The Gold Standard of Education: learning that transfers into life and the How X Works learning library.
Full chapter index
See the mechanism · Chapters 1–2
Build the worked model · Chapters 3–5
Test what changes · Chapters 6–8
Check evidence and limits · Chapters 9–10
Practise and continue · Chapters 11–12
A clone is the recovered descendants assigned to one founder under a stated labelling rule. Its competitive arena may be one crypt, one stem-cell niche, one blood compartment or a spatial field. Mixing arenas can make ordinary local replacement look like organism-wide expansion.
Write the outcome before fitting a model: frequency among labelled cells, absolute cell count, fraction of compartments occupied, or persistence across visits. Record when the label could first occur. An early founder has more time to grow than a late founder even with identical fitness.
Separate cell state from clone identity. A state can arise in several clones, and one clone can contain several states. This prevents an abundant state from being mistaken for a single expanding lineage.
Evidence check: for Define the clone and the competitive arena, preserve the observation before the interpretation. Record the original allele and the exact rescued phenotype. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Neutral drift means that lineages differ in realised success even when they have equal expected reproductive performance. In a fixed niche, one stem cell can replace a neighbour by chance; repeated events eventually extinguish some labels and enlarge others. The average labelled contribution may remain stable while the clone-size distribution widens.
The benchmark requires a replacement rule, effective population size, turnover rate and boundary. A well-mixed Moran-like model and a spatial neighbour-replacement model can have different fixation times and shapes. Choose the model from tissue organisation, not mathematical convenience alone.
Neutral does not mean nothing happens. It predicts stochastic expansion and loss. Selection is evidence that the observed dynamics are poorly explained by the neutral benchmark under defensible parameters.
Evidence check: for Build the neutral replacement benchmark, preserve the observation before the interpretation. Record reversion, same-gene change and second-site alternatives. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 3 OF 12 · Build the worked model
3. Preserve clone-size observations before fitness estimates
Start with a table of marker, count, region, time, assay quality and exclusions. Keep zeroes distinct from unobserved compartments. A clone absent from a later biopsy may be extinct, outside the biopsy, below detection or lost during processing.
Plot complementary cumulative clone-size distributions and the fraction of surviving clones. Means alone hide the simultaneous loss of many small clones and growth of a few large ones. Longitudinal data are stronger than a single terminal sample because they constrain the direction of change.
Do not discard the largest clone as an outlier merely because it changes the fit. Test contamination, collision and sample provenance first. If it is real, it is part of the biological distribution the model must explain.
Evidence check: for Preserve clone-size observations before fitness estimates, preserve the observation before the interpretation. Record matched parental controls and reciprocal crosses. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 4 OF 12 · Build the worked model
4. Work an expansion example with honest denominators
Imagine clone A rises from four to thirty-one recovered cells while clone B rises from five to eight. Clone A came from one densely sampled compartment; B was sought across all six. The raw ratio favours A, but the sampling opportunities differ.
Recalculate within matched compartments and report total cells recovered in each. If A is 31 of 200 cells in one compartment while B is 8 of 40 across six, neither percentage answers whole-tissue abundance without a sampling-weight model.
A defensible record includes the tissue volume, eligible cells, labelled cells, barcode-positive cells and assigned cells. Each denominator answers a different question and can change between visits.
| Clone | Cells sampled at T1 | Cells sampled at T2 | Compartments sampled |
|---|---|---|---|
| A | 4 | 31 | 1 of 6 |
| B | 5 | 8 | 6 of 6 |
| C | 2 | 0 | 6 of 6 |
| D | 8 | 14 | 3 of 6 |
Evidence check: for Work an expansion example with honest denominators, preserve the observation before the interpretation. Record joint genotype counts before interpreting ratios. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 5 OF 12 · Build the worked model
5. Estimate selection only within a declared model
A selection coefficient summarises a relative growth or replacement advantage under a specified model. It is not a direct property read from clone size. Mutation time, turnover and carrying capacity can trade off against the coefficient, so several parameter combinations may fit one terminal distribution.
Use likelihood or simulation to compare neutral and selective models, and show parameter intervals rather than only the best fit. Calibrate detection power with simulated clones passed through the observed sampling design. Weak selection may be indistinguishable from drift in a short record.
If the model assumes constant fitness, test whether a time-varying or compartment-specific advantage changes the conclusion. Biological context can alter competition without altering the inherited marker.
Evidence check: for Estimate selection only within a declared model, preserve the observation before the interpretation. Record dosage, penetrance and quantitative rescue. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 6 OF 12 · Test what changes
6. Separate birth, death, migration and labelling alternatives
A clone can grow because its cells divide faster, die less, differentiate less, enter the sampled compartment or were labelled more efficiently. Counts alone may not separate these mechanisms. Pair lineage data with proliferation, death, state and spatial measurements chosen before interpretation.
Migration is especially easy to overclaim. The same marker in two sites can reflect movement, a shared upstream founder or barcode collision. Temporal ordering and independent labels help discriminate them.
Perturbation can be useful, but it also changes the competitive arena. If treatment kills neighbouring cells, apparent clonal fitness may reflect released space rather than an intrinsic growth programme.
Evidence check: for Separate birth, death, migration and labelling alternatives, preserve the observation before the interpretation. Record informative meioses and every recombinant class. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 7 OF 12 · Test what changes
7. Respect spatial structure and niche boundaries
Competition is often local. A clone can fill one crypt while remaining rare across a colon, or dominate one marrow region while contributing modestly to blood. Spatial coordinates let the analysis distinguish local fixation from broad expansion.
Map clone boundaries, adjacency and tissue landmarks. Compare observed clusters with a spatial null model that preserves sampling density. A dispersed pattern and a contiguous patch can have the same cell count but imply different histories.
When sections are thin, a three-dimensional clone may appear as separated islands. Serial sections, volumetric imaging or carefully spaced samples can show whether gaps are biological or geometric.
Evidence check: for Respect spatial structure and niche boundaries, preserve the observation before the interpretation. Record flanking markers and donor-chromosome extent. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Sampling favours accessible, viable and assay-compatible cells. A dense biopsy can overrepresent one patch; dissociation can lose fragile states; barcode recovery can depend on expression or coverage. These processes reshape the observed clone-size distribution.
Define inclusion probabilities where possible and use stratified sampling across space and time. Technical replicates estimate recovery variability but do not replace biological replication. Rare clones need enough total observations to be distinguishable from dropout.
Compare several plausible sampling models. If a selection claim disappears under modest spatial weighting or detection correction, report it as sampling-sensitive rather than settled.
| Design | Cells recovered | Regions | Largest observed clone | Singleton fraction |
|---|---|---|---|---|
| Dense central biopsy | 400 | 1 | 84 | 0.12 |
| Stratified spatial sample | 400 | 8 | 29 | 0.31 |
Evidence check: for Model lineage-sampling bias explicitly, preserve the observation before the interpretation. Record background, temperature and blinded scoring. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Evidence strengthens when the same model explains independent tissues, animals or time points without retuning every parameter. Posterior predictive checks or simulated summaries should reproduce survival fractions, size distributions and spatial clustering, not just one headline statistic.
Inspect residuals by compartment and time. Systematic failure can reveal a missing boundary, a burst of labelling or state-dependent capture. A more complex model earns its place only if it explains a relevant discrepancy and remains identifiable.
Predefine which comparisons constitute replication. Multiple sections from one clone-rich tissue are not independent biological replicates.
Evidence check: for Use replication and model criticism, preserve the observation before the interpretation. Record a molecular readout that separates bypass from repair. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 10 OF 12 · Check evidence and limits
10. Know the limits and common misconceptions
Large does not mean selected; persistent does not mean immortal; neutral does not mean static. A marker associated with a clone may be a passenger rather than its cause. Clone abundance in a sample is not whole-organ abundance unless sampling supports that translation.
Selection can change over time and environment. A clone advantageous during injury may be neutral afterward. Conversely, a small clone can carry strong fitness if it arose late.
Human tissues often provide sparse, opportunistic and cross-sectional evidence. In that setting, use restrained language: compatible with positive selection, inconsistent with the fitted neutral model, or unresolved between timing and fitness.
Evidence check: for Know the limits and common misconceptions, preserve the observation before the interpretation. Record independent recreation and reversal of the candidate allele. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 11 OF 12 · Practise and continue
11. Independent practice and changed-context transfer
Practice case: one clone forms 35% of a central biopsy but appears in none of four peripheral biopsies. Construct two explanations—local selective expansion and neutral local fixation—and name a measurement that separates them.
Changed context: a second sampling round is spatially stratified and finds the clone is 6% tissue-wide. Recalculate which claim survives. Then suppose the marker first appeared one month before sampling rather than one year earlier; explain how the same size changes the implied growth advantage.
Finish with a three-sentence record: the direct observation, the model-dependent inference and the sampling or timing condition most likely to reverse it.
Evidence check: for Independent practice and changed-context transfer, preserve the observation before the interpretation. Record the observation that would falsify the current map. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Does a very large clone prove positive selection?
No. Its evidential strength depends on mutation timing, neutral turnover, space, founder number and sampling. A fitted neutral alternative and a matched sampling model are needed.
Can neutral drift produce fixation?
Yes. In a finite compartment, stochastic replacement can eliminate other lineages and leave one lineage occupying the compartment without a fitness difference.
Why keep spatial information?
Because local fixation, migration and tissue-wide expansion can produce similar aggregate counts but different maps.
Continue with clonal lineage tracing, barcodes and convergent histories and mosaic analysis and cell autonomy.
Authoritative background includes the NIH-hosted review Lineage tracing in human tissues, a stem-cell neutral-drift study testing crypt lineage dynamics, and research on how spatial structure changes detection of selection and drift. Sources and links were checked on 22 October 2026.
Evidence check: for FAQ, sources and connected reading, preserve the observation before the interpretation. Record the smallest next cross that resolves a live alternative. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.