MCAT content library / Research & data / Reading a multi-figure experiment from claim to conclusion
Keep the question, readout, and control attached to each result before combining an experiment’s panels.
About 17 minutes, including the worked examples and checks.
From a dataset to a supported claim · Lesson 6 of 6 · Continue in this path
Useful foundations
On this page
- Concept: each panel answers a narrower question than the paper’s title
- Pattern: one compound changes two readouts differently
- Rule: test an assay explanation before treating it as biology
- Application: use a rescue to locate a functional limitation
- Put the panels together without erasing their differences
- Check your reasoning
- Continue learning
- Sources
Concept: each panel answers a narrower question than the paper’s title
An experiment can combine a cell count, a reporter signal, an enzyme assay, and a rescue condition. Those measurements are not interchangeable votes for one conclusion. A cell count addresses how many cells were counted; a reporter addresses a measured signal; an activity assay addresses a reaction under its assay conditions. The task is to connect the evidence without quietly substituting one quantity for another.
For each panel, recover the manipulated variable, comparison, measured quantity, denominator, and time point. Then write the narrow observation before proposing its explanation. AAMC’s research and data skills include comparing hypotheses, using controls, combining evidence, and recognizing limits on inference.12
A useful working distinction is observation → interpretation → next discriminating test. “The reporter signal is lower” is an observation. “The cells divide more slowly” is an interpretation. A measurement that tracks divisions can test that interpretation more directly than another reading of the same reporter.
Pattern: one compound changes two readouts differently
The following study is original and illustrative. Equal cultures begin with 0.50 million viable cells. They receive vehicle or compound D for 24 hours. Direct viable-cell counting is independently validated and is not affected by D. The fluorescence reporter is sometimes used as a proxy for cell number, but that relationship has not yet been established in the presence of D.
| Condition | Viable cells (millions) | Total reporter fluorescence (AU) |
|---|---|---|
| Vehicle | 2.00 | 100 |
| Compound D | 1.00 | 25 |
The D culture has half as many viable cells at the endpoint, but one quarter of the total reporter signal. Both cultures increased above their 0.50-million starting count. Therefore “D eliminated 75% of the cells” is not supported by these endpoints. A smaller endpoint count can reflect slower division, increased loss, or a combination; it is not a direct count of deaths.
Normalize the reporter by the counted cells. Vehicle gives 100/2 = 50 AU per million cells; D gives 25/1 = 25. Even after accounting for cell number, the reporter is lower. That residual difference could reflect a changed cellular state, direct interference with the reporter, or both.
Rule: test an assay explanation before treating it as biology
Panel B holds the amount of fluorescent reporter fixed in a cell-free preparation. D and vehicle are added at the same relevant concentrations used for the fluorescence reading. The detector and all other conditions are matched.
| Condition | Reporter amount | Fluorescence (AU) |
|---|---|---|
| Vehicle | Fixed and equal | 100 |
| Compound D | Fixed and equal | 50 |
The signal decreases in a preparation with no cells to divide or die. This supports direct interference with the fluorescence readout under the tested conditions. It explains why an uncorrected fluorescence value is not a reliable stand-in for cell number here. It does not erase Panel A’s independently counted cell-number difference.
Nor does this control prove that every change in cellular reporter signal is caused only by interference. The intracellular environment can differ from the cell-free preparation. Treat the control as evidence about a plausible measurement mechanism, and retain the independent readout rather than overextending a correction.
Application: use a rescue to locate a functional limitation
The researchers next study a hypothetical growth pathway. Enzyme E supplies a product that can also be supplied downstream by metabolite M. The passage stipulates that M can enter this pathway after E’s step. Enzyme activity is measured per microgram of protein by an independent method that separates and quantifies product; validation shows that D and M do not interfere with product detection.
| Condition | E activity (units/µg protein) | Viable cells at 24 h (millions) |
|---|---|---|
| Vehicle | 10 | 2.00 |
| Compound D | 4 | 1.00 |
| D + metabolite M | 4 | 1.90 |
| Vehicle + M | 10 | 2.00 |
D is associated with lower measured E activity. Adding M in the presence of D restores the endpoint count close to the vehicle value while the measured E activity remains low. Under the stated pathway model, this pattern supports a limitation that can be bypassed downstream of E’s step. The vehicle-plus-M condition helps show that M does not simply raise the endpoint count above the untreated level in these observations.
This is a more specific argument than “the rescue proves D binds E.” The activity result could arise through several mechanisms, and the supplement can compensate for a limitation without identifying D’s direct molecular target. Direct-target claims require evidence directed at that molecular interaction. The combined panels support a functional explanation at a bounded level.
Put the panels together without erasing their differences
A defensible synthesis has three parts. First, D lowers the independently counted 24-hour cell number relative to vehicle. Second, D also affects the reporter assay, so fluorescence overstates the reduction in counted cells if treated as a simple cell-number proxy. Third, the activity and supplementation results support a downstream-bypass explanation for part of the growth limitation under the supplied pathway assumptions.
The evidence does not distinguish slower division from increased cell loss, identify a binding site, show that E is D’s only target, or provide a statistical significance result. No uncertainty estimate is supplied. Describe the stated pattern and its mechanistic compatibility without adding a test that was never reported.
The best next experiment depends on which uncertainty matters. To distinguish division from cell loss, track individual cells over time and record both events. To investigate a direct effect on E, use an appropriately controlled purified-enzyme experiment and complementary target evidence. To test whether the reporter remains useful, validate its relationship to an independent count across the conditions where it will be used.
Do not choose an experiment simply because it is more elaborate. Choose the one whose possible outcomes would separate the explanations still compatible with the present results. A new panel adds value when it resolves a particular ambiguity, not merely when it repeats the same pattern with a different label.
ACTIVE PRACTICE · COMPARE AND TRANSFER
Combine Count, Reporter, and Rescue Without Inventing a Target
The original 24-hour matched cultures start with 0.50 million viable cells. Vehicle ends with 2.00 million and reporter 100 AU; compound D ends with 1.00 million and reporter 25 AU. An independent assay shows E activity 10 units/µg in vehicle and 4 under D; adding downstream metabolite M leaves E at 4 but restores viable count to 1.90 million.
Examine the original evidence
Results are bounded to the original supplied observations. Instructional checks do not establish assessment calibration or clinical effectiveness.
Check your reasoning
Choose an answer before opening its feedback. Each option explains the reasoning behind that choice.
Check 1
Two cultures each begin with 0.40 million cells. At 24 hours, a validated independent count gives 0.80 million for vehicle and 1.20 million for compound Q. Total reporter signals are 80 and 60 AU, respectively. Which interpretation is supported?
A. Q decreased the counted endpoint population because its reporter signal was lower.
Review this choice. The independent counts give the opposite result:1.20 million is greater than 0.80 million. A lower reporter cannot replace those counts.
B. Q increased the counted endpoint population while decreasing reporter signal per counted cell.
Correct. The endpoint count rises by 50%, but reporter per million cells falls from 80/0.80 = 100 to 60/1.20 = 50 AU. The two readouts change in different directions.
C. Q increased reporter signal per counted cell by 50%.
Review this choice. The calculated per-cell reporter falls by 50%; it does not rise. The 50% increase applies to the endpoint cell count.
D. The reporter readings prove that every Q-treated cell has stopped dividing.
Review this choice. Neither total fluorescence nor an endpoint population provides a record of every individual division. This claim is also inconsistent with treating the larger count as evidence of complete arrest.
The principle to carry forward
Calculate each readout in its own units before deciding whether it supports the same biological story.
Check 2
In Panel A, the vehicle and D cultures begin at the same count and end at 2.00 and 1.00 million viable cells. Which additional observation most directly distinguishes slower division from greater cell loss as explanations for the lower D endpoint?
A. A second endpoint measurement using the same fluorescence reporter.
Review this choice. The reporter has a demonstrated interference problem and still does not distinguish the underlying population events.
B. An enzyme-activity measurement from the original endpoint samples.
Review this choice. An activity difference may suggest a mechanism, but it does not directly count division and loss over time.
C. Individual-cell tracking that records division and loss events throughout the interval.
Correct. The competing explanations concern different events in the population history. Tracking both events directly addresses that distinction and can reveal a mixture of mechanisms.
D. The compound concentration measured only at the start of treatment.
Review this choice. Initial concentration characterizes exposure, not how the cell population changed over the interval.
The principle to carry forward
Match the next measurement to the unresolved mechanism, not merely to another correlating endpoint.
Continue learning
From a dataset to a supported claim
Lesson 6 of 6. You have reached the end of this path.
- Previous in From a dataset to a supported claim: Statistical significance and effect size: what the result means
- Return to this path’s overview
See every step in From a dataset to a supported claim
From a dataset to a supported claim
Interpret variability and uncertainty before combining evidence from multiple figures.
- Reading graphs: axes, scales, and the comparison that matters
- Mean, median, and spread: describing a dataset
- Accuracy, precision, and sources of measurement error
- Confidence intervals, error bars, and sample size
- Statistical significance and effect size: what the result means
- Reading a multi-figure experiment from claim to conclusion · You are here
Related lessons
- Experimental Controls: What a Comparison Can Show
- Hypotheses, predictions, and competing explanations
- Normalized data, percent change, and fold change
- Statistical significance and effect size: what the result means
- Measuring gene expression and testing gene function
Browse research & data · MCAT content library
MCAT preparation and roadmap · Aim High, Score Higher