The Tumor Growth tab provides five statistical modeling approaches selectable from the Statistical Model dropdown:

• Linear Mixed-Effects Model (LME4) — fits a mixed-effects model to repeated tumor volume measurements. Accounts for inter-animal variability and unequal observation times. Recommended for most longitudinal datasets.
• GAMM (smooth time × treatment) — fits a group-specific smoother on Day via gamm4. Use this instead of a high-order polynomial when growth is visibly non-linear: it estimates the shape from the data rather than assuming log-linearity. Diagnostics include a basis-dimension (k) check and a concurvity report.
• Area Under Curve (AUC) — summarises each animal’s total tumor burden as the trapezoidal AUC of its growth curve. Robust to missing timepoints, but discards trajectory shape information.
• Bayesian LMM (brms) — the LME4 model fitted by MCMC, returning full posterior distributions instead of point estimates and p-values.
• Bayesian GAMM (brms + smoothers) — the GAMM equivalent, with the smoother fitted in Stan.

R14.3: this tab does not report Tumor Growth Inhibition. TGI is an endpoint summary — a ratio of two group means on one day — while every model here estimates a growth trajectory , and collapsing a trajectory to a single-day ratio discards most of what was measured. TGI is computed where it belongs to the question being asked: the Toxicity tab (as the efficacy half of the therapeutic window) and the Drug Synergy tab (as the fractional effect the Bliss model acts on). Both expose an endpoint-estimand selector, because a ratio of endpoint means is exactly where early removals bias the answer.

Model

A linear mixed-effects model is fitted to transformed tumor volumes:

log(Volume) ~ Treatment × Day + (1 | ID)

The Treatment × Day interaction captures whether treatment groups differ in their growth trajectories over time, not just at a single endpoint.

Transformation

Log (default) stabilises variance and linearises exponential growth. Square root suits moderate heteroscedasticity. Use 'None' only for already-linear data.

Random Effects

'Intercept only' (default) allows each animal its own baseline volume. 'Intercept + Slope' additionally allows individual growth rates — more flexible but may fail to converge with small groups. 'None' drops the animal-level term entirely and is appropriate only when each animal contributes a single observation, since it otherwise treats repeated measures as independent.

Cage Effect

Auto (default) classifies the housing design and applies the matching treatment, because the correct handling depends on the design rather than on preference:

• Crossed — a cage holds more than one treatment. Cage is separable from treatment and enters as a random intercept.
• Nested, replicated — one treatment per cage, several cages per arm. Cage nests within treatment as a random effect; ignoring it treats cagemates as independent and understates standard errors.
• Nested, confounded — one cage per treatment. Cage and treatment are the same variable and no model can separate them. The result reports the effect as inseparable rather than silently crediting it to treatment.

The remaining options force a specific handling. Include if not collinear (legacy) is the pre-v0.12 behaviour, kept so earlier analyses can be reproduced.

ANOVA Tab

Wald chi-square tests for Treatment, Day, and the Treatment × Day interaction, reported as Chisq / Df / Pr(>Chisq) . A significant interaction indicates that treatment groups differ in their growth trajectories — the primary analysis target.

These are chi-square rather than F tests because the model is fitted by maximum likelihood: a mixed model has no exact residual degrees of freedom, so an F reference distribution would need an approximation the chi-square avoids. The trade-off is that it is anti-conservative in small samples — with few animals per arm, prefer the randomisation test below.

Comparisons Tab

Pairwise group comparisons at the mean observation day using estimated marginal means (EMMs). Two independent settings control this, and conflating them is the usual mistake:

Post-hoc Comparisons

Which comparisons are made — vs Reference Group (each treatment against the reference only) or All Comparisons (every pair). This defines the family.

Correction

How that family is corrected: Dunnett (exact, multivariate-t), Tukey HSD, Bonferroni, Holm, Benjamini-Hochberg FDR, or None. Dunnett is the default and is the right partner for a vs-reference family — it accounts for the correlation between comparisons that share a control arm, which Bonferroni ignores and therefore over-corrects.

The correction is applied over exactly the set of comparisons requested, and the table header states which correction was actually used rather than asserting one. Comparisons are marginalised at the mean study day; for differences in trajectory across time, see the Treatment Effects tab.

Randomisation Test

An optional exact test of the Treatment × Day interaction that avoids both the denominator-degrees-of-freedom approximation and the assumption of normally distributed random effects. Treatment labels are permuted and the model refitted; the p-value is the proportion of permutations giving a statistic at least as extreme as the observed one.

The panel reports the number of exchangeable units, how many permutations converged, whether the permutation set was enumerated exhaustively, and the smallest p-value the design can attain. With few units that floor can exceed 0.05, in which case no amount of signal can produce a significant result — and the asymptotic p-value beside it should not be trusted.

Treatment Effects Tab

Estimated marginal means (EMMs) of transformed volume per treatment group. EMMs are shown at the mean observation day and at five quantile days (min, Q1, median, Q3, max), capturing the Treatment × Day interaction that a single mean-day estimate would miss.

Growth Rates Tab

Per-animal exponential growth rates estimated by log-linear regression of each animal’s volume trajectory independently. Reported as growth rate (log-scale slope), fold change, and % change per day with 95 % CIs and R². These are descriptive; for group-level inference use the ANOVA and Comparisons tabs.

Diagnostics Tab

Four residual plots, each answering a different question about whether the model's assumptions hold — which is what makes the p-values above mean anything:

Q–Q plot

Are residuals normally distributed? Systematic curvature at the tails suggests the transformation choice should be revisited.

Residuals vs fitted

Is the mean structure right? Curvature means the log-linear growth assumption is wrong — try GAMM.

Scale–location

Is the variance constant on the modelling scale? A rising trend means heteroscedasticity survived the transformation.

Random-effects Q–Q

Are the per-animal effects themselves normal? This is a separate assumption from residual normality and the one most often overlooked. If it fails, prefer the randomisation test, which assumes neither.

Influence measures (Cook's distance, DFBETAS) are also available: they identify individual animals whose removal would materially change the conclusion. A result resting on one mouse is worth knowing about before publication rather than after.

Fits a generalised additive mixed model via gamm4, replacing the linear Day term with a group-specific smoother:

log(Volume) ~ Treatment + s(Day, by = Treatment) + (1 | ID)

Use this when growth is visibly non-linear — plateauing, regression-then-regrowth, or a delayed onset of effect — where forcing a single slope per arm would average across distinct phases and understate the treatment effect. The smoother estimates the shape from the data instead of assuming it, which is the advantage over adding polynomial terms.

Basis dimension (k) check

Tests whether the smoother had enough flexibility to capture the trend. A significant result means k was too small and the fitted curve is over-smoothed — the shape you are interpreting is partly an artefact of the basis.

Concurvity

The smooth analogue of collinearity: whether one smoother can be approximated by another. High concurvity makes individual smooth terms unidentifiable, so their separate interpretation is unsafe even when the overall fit is good.

Comparisons on this path are per-day contrasts rather than a single slope difference, because a smoother has no single slope. The Comparisons header states this explicitly.

Trapezoidal AUC of tumor volume over time is computed per animal. AUC (mm³·day) captures total tumor burden integrated over the study, not just the final timepoint.

Individual AUC

Per-animal AUC values with group summaries (mean, SD, SEM).

AUC Summary

An omnibus Welch one-way ANOVA across groups, followed by the pairwise comparisons and correction selected in the sidebar. Welch is used rather than the classical ANOVA because AUC variance almost always scales with the mean — a control arm that grows unchecked is far more variable than a responding one — so the equal-variance assumption is violated in exactly the comparison of interest.

Variance check

A formal test of equal variances is reported beside the omnibus result, so the assumption is visible rather than assumed. Welch already tolerates unequal variances; the test tells you how far from equal they were.

Extrapolation no longer feeds this calculation. Appending an imputed observation and treating it as measured data made standard errors and p-values ignore the imputation, understating uncertainty. The Extrapolation Points control now affects the growth-trajectory plots only.

Method

Two Bayesian modes are available: Bayesian LMM fits the same formula as LME4, and Bayesian GAMM fits the GAMM formula with the smoother estimated in Stan. Both use full Bayesian inference via brms / Stan. Instead of point estimates and p-values, every parameter has a posterior distribution : a probability distribution over plausible values given the data and priors.

log(Volume) ~ Treatment × Day + (1 | ID)

Results are summarised as posterior means with 95 % credible intervals. A 95 % CrI has a direct probability interpretation: there is a 95 % posterior probability that the true parameter lies within the interval — unlike a frequentist confidence interval, which is a statement about the procedure rather than the parameter.

R14.7: two different interval types appear under the same Lower_95_CrI / Upper_95_CrI headings, and it is worth knowing which you are reading. The Summary tab reports raw model parameters straight from brms: an equal-tailed interval, cutting 2.5 % from each tail. The Treatment Effects and Comparisons tabs come from emmeans, which returns a genuine highest-density interval — the shortest interval containing 95 % of the posterior. The two coincide when a posterior is symmetric and diverge when it is skewed, which is exactly the case for variance components and for the back-transformed time ratios in the Bayesian survival model.

Prior Strength

Priors express what you believe about parameter values before seeing the data. They regularise estimates and prevent the model from chasing noise in small samples. All presets use Normal priors on fixed effects and Exponential priors on variance components; only the widths differ.

Skeptical

Default. N(0, 0.25) on fixed effects; Exponential(2) on SDs. Encodes the reasonable belief that treatment effects are small — effects > ±0.5 log-units per day are considered very unlikely a priori. Recommended for exploratory preclinical studies where large effects have not been previously established.

Weakly informative

N(0, 1) on fixed effects; Exponential(1) on SDs. Broader than Skeptical — appropriate when prior work suggests the treatment category can produce moderate-to-large effects.

Informative

N(0, 0.5) on fixed effects; Exponential(2) on SDs. Between Skeptical and Weakly informative. Useful when effect sizes in the 0.1–0.5 log-unit range are plausible based on prior data.

Diffuse

N(0, 2.5) on fixed effects; Exponential(0.5) on SDs. Near-flat priors; use with caution on small datasets as estimates can be driven entirely by noisy data.

Manual

Specify the prior family and parameters for each model class independently. Fixed effects (b): Normal or Student-t. Intercept: same. Random-effect SD and residual sigma: Exponential or half-Normal.

Result Tabs

• Summary — posterior median and 95 % CrI for every model parameter.
• Treatment Effects — group-level estimated marginal means (posterior) at the mean study day.
• Comparisons — posterior contrasts of each group vs the reference.
• Posterior Check — overlay of observed data density vs draws from the posterior predictive distribution. Systematic mismatch indicates model mis-specification. A coverage plot reports whether the stated credible intervals actually contain the observed values at the claimed rate.
• Diagnostics (Bayesian view) — Rhat and effective sample size (ESS) per parameter, with trace and rank plots. Rhat > 1.01 indicates the chains have not mixed; low ESS means the posterior summaries are themselves noisy. Rank plots detect the mixing failures that trace plots hide once chains overlap.
• Influence — Pareto-k values from PSIS leave-one-out. Values above 0.7 mean the importance sampling is unreliable for that observation, which usually flags an animal the model fits badly rather than a coding error.
• Growth Rates — per-animal log-linear regression rates (same as LME4, not MCMC-based).

When a Necrotic Flag column is mapped in Data Upload, a handling selector appears in the sidebar. Necrotic tumors have a non-viable core that inflates caliper-based volume estimates without contributing to proliferating tissue.

Exclude (recommended)

Removes flagged observations from the model input. Conservative and publication-defensible. The Necrosis tab reports how many observations were flagged and which groups were affected, regardless of handling mode.

Covariate

Adds a 0/1 .necrotic_flag fixed effect to the LMM formula. Estimates the average volume bias attributable to necrosis. Useful when necrosis is approximately evenly distributed across groups and you want to retain all timepoints.

No adjustment

All observations are used as-is (current default when no necrotic column is mapped). The Necrosis summary table is still shown for documentation purposes.

Reference Group

The vehicle/control group used as comparator. Auto-detected from common names (Vehicle, Control, DMSO, PBS, Saline) or defaults to the first group alphabetically.

Post-hoc Comparisons

Which family of comparisons to make — vs Reference Group (default) or All Comparisons. Paired with the separate Correction setting.

Correction

How the chosen family is corrected: Dunnett (exact multivariate-t, default), Tukey HSD, Bonferroni, Holm, Benjamini-Hochberg FDR, or None.

Extrapolation Points

Affects the growth-trajectory plots only. Number of recent timepoints used to estimate a slope when projecting the curve for animals that left the study early; projected segments are drawn dashed. Set to 0 to disable. This no longer feeds any statistical model — treating an imputed point as measured data made standard errors ignore the imputation.

Time Column

Numeric study day, or calculated from a date column mapped in Data Upload (days elapsed from first observation).

The Survival tab fits Kaplan–Meier (KM) survival curves and a Cox proportional hazards model to time-to-endpoint data. In mouse tumor experiments the 'endpoint' is typically the day a tumor exceeds the IACUC volume limit, not true animal death.

Three outputs are produced:

• Kaplan–Meier curves with log-rank test p-values.
• Cox proportional hazards model reporting hazard ratios (HR) per group relative to the reference.
• Risk table showing animals at risk, events, and KM estimates at each event time.

The KM estimator computes a non-parametric step-function survival probability S(t) = P(T > t) from event times. Each downward step corresponds to one or more events on that day.

Log-rank test

Tests whether the survival curves of two or more groups are statistically equivalent. The omnibus p-value tests all groups simultaneously; pairwise p-values are computed by restricting each comparison to the relevant two groups (df = 1).

Median survival

Day by which 50 % of animals reached the endpoint. 'Not reached' if fewer than half had events before study end.

Censoring

Animals that never reach the endpoint contribute their last observation day as a censored time. This correctly uses partial information rather than excluding them.

When all animals in a group are censored (zero events), a log-rank test is not meaningful. The model falls back to pairwise comparisons using only groups with at least one event.

The Cox model estimates the hazard ratio (HR) for each treatment group relative to the reference:

h(t) = h₀(t) × exp(β₁ × Treatment)

HR < 1

Lower hazard than reference — animals reach the endpoint more slowly (better outcome).

HR = 1

Identical hazard to reference.

HR > 1

Higher hazard than reference — animals reach the endpoint faster (worse outcome).

Firth's bias-reduced estimation is applied automatically when any group has zero events. Standard Cox partial likelihood produces infinite coefficient estimates in this case; Firth's penalisation regularises them.

The proportional hazards assumption requires that the HR is constant over time. Crossing KM curves indicate a time-varying HR, which violates this assumption — interpret Cox results with caution in that case.

Mapped censor column

When an Event/Censor column is mapped in Data Upload (0 = censored, 1 = event), it is used directly.

Auto-detection

When no censor column is mapped, animals that disappear before the final study day are automatically treated as events; animals present at the final day are censored.

Volume threshold

Optionally generate an event indicator by flagging animals whose tumor volume exceeds a specified threshold at any timepoint.

Natural death

If checked (recommended), animals that die before reaching the volume threshold are counted as events. If unchecked, only volume threshold crossings are events and natural deaths are censored.

The Toxicity tab provides two complementary analysis modes selectable from the Statistical Model dropdown:

• Body Weight Analysis — models longitudinal weight trajectories using a linear mixed-effects model, and derives AUC-based and time-to-event (KM) toxicity summaries.
• Efficacy-Toxicity Analysis — integrates body weight toxicity with tumor efficacy data to assess the overall therapeutic window. Requires a tumor volume column.

Body weight is used as the toxicity surrogate throughout. When 'Subtract estimated tumor weight' is enabled, the gross weight is corrected by subtracting Volume(mm³) × density(g/cm³) / 1000, giving a net host weight.

Mixed-Effects Model

Fits a linear mixed-effects model to all repeated weight measurements:

Weight ~ Treatment × Day + [covariates] + (Day | Mouse)

Random slopes allow each mouse's weight trend to deviate from the group average. Fixed effects test whether trajectories differ by treatment group over time.

REML

Restricted Maximum Likelihood — preferred for final parameter estimates (default).

ML

Maximum Likelihood — use when comparing models with different fixed-effect structures.

Covariates (optional):

• Tumor Volume — controls for tumor-driven weight changes independent of drug toxicity.
• Sex — adjusts for sex-specific baseline weight differences.
• Initial Mass — adjusts for pre-treatment weight variation across animals.

Body Weight AUC

For each mouse, body weight is expressed as % change from baseline at each timepoint, then a trapezoidal AUC is computed over the study. A negative AUC indicates sustained weight loss; a positive AUC indicates net weight gain.

• Individual AUC values are shown as a jitter plot with a configurable centre mark (mean or median) and error bars (SEM, SD, or 95% CI).
• Units are % · days : e.g. a value of −100 means an average 5% weight loss sustained over 20 days.
• Pairwise group comparisons use t-tests with Benjamini-Hochberg correction.

Nadir Analysis

The nadir (minimum weight) is identified per mouse. Reported metrics include nadir day, nadir weight, % loss at nadir, and days to recovery.

Weight Loss KM Curve

Treats each mouse as an event-time observation where the event is crossing the weight-loss threshold (default 20%). Mice that never reach the threshold are censored at their last observation. A Kaplan-Meier curve is fitted per group; group differences are tested with the log-rank test.

Weight Loss Threshold: The % body weight loss that constitutes a toxicity event. The default (20%) is a common IACUC endpoint; adjust to match study-specific welfare criteria.

Efficacy Metrics

Three metrics quantify anti-tumor efficacy per mouse relative to the control group:

Final TGI (%)

Tumor Growth Inhibition at the final timepoint: TGI = (1 − V_treated / V̅_control) × 100 . Values > 0 indicate tumor suppression; > 100 indicates tumor regression.

Tumor AUC

Trapezoidal AUC of tumor volume over time, normalised to the control AUC: (1 − AUC_treated / AUC_control) × 100 . Captures the full trajectory rather than a single endpoint.

Therapeutic Window Metric (TWM)

A summary ratio combining efficacy and toxicity per group:

TWM = TGI(%) / Mean Body Weight Loss(%)

Higher TWM indicates a more favourable balance. The denominator is the group mean of per-mouse maximum weight loss (not the single worst-case mouse). A noise floor prevents division by near-zero weight changes. Displayed as a ranked bar chart.

Ranking bounds

TWM is a ratio of two estimated quantities, so its uncertainty is wider than either component's and does not follow a normal distribution. Bootstrap intervals are reported and the ranking note states whether the top arm is genuinely separated from the second or whether their intervals overlap. A bar chart ordered by point estimate looks like a ranking whether or not the data support one.

Bootstrap resamples

Number of mouse-level resamples used for those intervals. The control arm is resampled too, so uncertainty in the TGI denominator propagates rather than being treated as known. Set to 0 to skip, which returns point estimates with no interval.

Endpoint estimand

Model (default) estimates group means at the endpoint from the fitted model, using every observation each animal contributed. Observed uses only animals still present at the endpoint. The two differ whenever animals are removed early, and the difference is not small: because the fastest-growing controls are the ones that leave, the observed estimand flatters the control arm and understates TGI. Per-arm attrition is reported beneath the table so the size of the issue is visible.

Safety-Efficacy Scatter

A per-mouse bivariate plot where each point represents one animal. X-axis shows the chosen efficacy metric; Y-axis shows maximum % body weight loss. Points in the top-right quadrant (high efficacy, high toxicity) represent the least favourable profile.

• Group ellipses — 68% confidence ellipses (1 SD) per treatment group.
• Group centroids — mean efficacy and mean max weight loss per group, shown as large diamonds.

Total Benefit Score

An integrated scalar combining efficacy and toxicity AUCs over time:

B = AUC(efficacy) − λ × AUC(toxicity)

A positive score indicates net benefit; negative indicates net harm. The trade-off parameter λ (lambda) controls how strongly toxicity is penalised:

• λ = 1.0 (default) — equal weight on efficacy and toxicity.
• λ > 1 — penalise toxicity more heavily; use for fragile populations or curative intent.
• λ < 1 — tolerate more toxicity; use when efficacy improvement is the primary goal.

Weight-Corrected TGI

Recalculates TGI using only mice that stayed within the safety weight-loss threshold throughout the study. Mice exceeding the threshold are excluded from the corrected calculation.

This prevents a drug that arrested tumor growth due to severe toxicity from appearing falsely effective. Comparing the standard and corrected TGI reveals the extent of this confound.

Safety threshold: Mirrors the Weight Loss Threshold in Body Weight Analysis. Mice whose maximum weight loss exceeds this value are excluded from the corrected TGI.

Reference Group

The vehicle/control group used as the denominator for TGI and AUC ratios. Auto-detected from common names (DMSO, Vehicle, Control, PBS, Saline) or defaults to the first group alphabetically.

Subtract Tumor Weight

Removes the estimated tumor mass (Volume × density / 1000) from gross body weight before all calculations, so that body weight reflects host tissue rather than tumor burden.

Tumor Density

Used only when tumor weight adjustment is enabled. Default 1.0 g/cm³ (water density). Typical soft-tissue tumor density is 0.9–1.1 g/cm³.

Baseline Day

The study day used to compute % change from baseline. Defaults to the earliest available day.

The Dose-Response tab models how tumor volume (or TGI) varies across dose levels. Three complementary analyses are performed simultaneously:

• Linear regression — tests whether there is a significant linear trend in tumor volume with increasing dose.
• One-way ANOVA — tests whether any dose group differs from the others, followed by Tukey HSD post-hoc comparisons if significant.
• Non-linear dose-response model (4PL) — fits a four-parameter log-logistic (Hill) model to quantify ED50 and curve steepness.

Tumor Growth Analysis must be run before using this tab — dose-response uses volume data preprocessed there. A Dose column must also be mapped in Data Upload.

Fits Volume ~ Dose by ordinary least squares. The slope quantifies the change in mean tumor volume per unit increase in dose.

Slope

Change in volume per dose unit. A negative slope indicates dose-dependent tumor suppression.

R²

Proportion of variance in volume explained by dose. Low R² with a significant slope can occur when individual variability is high relative to group differences.

p-value

Significance of the slope. A non-significant p-value does not exclude a non-linear relationship — inspect the 4PL model output and plots as well.

Treats dose as a categorical factor and tests whether group means differ. A significant F-test (p < 0.05) triggers Tukey HSD post-hoc comparisons to identify which specific dose pairs differ.

F-statistic

Ratio of between-group to within-group variance. Larger values indicate greater separation between dose groups relative to within-group noise.

Tukey HSD

Family-wise error rate corrected pairwise differences. Values in 'p adj' below 0.05 indicate a significant difference at the 5 % level after correction for all comparisons.

ANOVA assumes equal variances across groups (homoscedasticity). When this is violated, interpret alongside the linear regression, which is more robust to mild heteroscedasticity.

A four-parameter log-logistic (4PL / Hill) model is fitted via the drc package:

Volume = c + (d − c) / (1 + (Dose / ED50)^b)

b

Hill coefficient: steepness of the dose-response curve. Larger absolute values indicate a sharper transition.

c

Lower asymptote (response at saturating inhibitory dose).

d

Upper asymptote (response in the absence of drug).

ED50

Dose producing 50 % of the maximal effect. The primary pharmacodynamic summary statistic.

A dose-response study usually asks whether response is monotone in dose — not merely whether the groups differ somewhere. Three tests address that directly:

Jonckheere–Terpstra

The test that actually respects dose ordering. It is sensitive to a consistent monotone trend and, unlike ANOVA, gains power from the ordering rather than discarding it. This is the primary trend result.

Linear trend

p-value for a linear term in dose. Significant means a straight-line component is present; it says nothing about whether the relationship is only linear.

Quadratic trend

Detects curvature — a plateau or a reversal at high dose. A significant quadratic term alongside a significant linear one usually means the response saturates, which matters when extrapolating to untested doses.

A Bayesian dose-response mode is also available, fitting the same curve by MCMC and returning posterior distributions for ED50 and the slope rather than point estimates.

The Drug Synergy tab assesses whether a drug combination produces greater tumor suppression than predicted from the individual drugs acting independently. Two reference models are evaluated simultaneously:

• Bliss Independence — treats the two drugs as probabilistically independent. The expected combined TGI is the sum of individual TGIs, capped at 100 %.

Four treatment groups must be assigned in the sidebar: Control, Drug A only, Drug B only, and the Combination (A + B). Analysis can be run at a single timepoint or tracked over the full study duration.

Tumor Growth Analysis must be run first — Drug Synergy uses preprocessed volume data from that tab.

Under Bliss independence, drugs act on independent targets and their joint fractional effect (FE) is:

Expected TGI = min(FE_A + FE_B, 1.0) × 100

where FE = TGI / 100 for each single agent. The cap at 1.0 prevents the expected value from exceeding the biological ceiling.

Bliss Difference

TGI_combo − Expected TGI. Positive = synergy; negative = antagonism; near zero = independence.

The synergy verdict is a categorical label derived from three group means. Without an interval, a “Strong Synergy” resting on noise is indistinguishable on screen from one resting on a real effect.

Bootstrap resamples

Mouse-level resampling produces confidence intervals for the Bliss metrics. The control arm is resampled too, so uncertainty in the denominator propagates rather than being treated as known — which matters because every metric here is a ratio against control. Set to 0 to skip; the verdict is then a point estimate with no stated uncertainty.

Interval callout

Beneath the verdict, an explicit statement of whether the 95 % interval for Bliss excess excludes zero. This is the single most useful line in the tab: a verdict whose interval spans zero is not evidence of synergy, whatever the label says.

Endpoint estimand

Model (default) estimates endpoint group means from the fitted model, using every observation each animal contributed. Observed uses only animals still present at the endpoint. Because the fastest-growing controls are the ones removed early, the observed estimand shrinks the control mean and inflates every fractional effect derived from it — which propagates directly into both Bliss and the Bliss expectation. Per-arm attrition is shown so the exposure is visible.

A Bayesian synergy mode is also available: it propagates posterior uncertainty from the growth model through the Bliss calculation, giving credible intervals for the metrics directly rather than by resampling.

When 'Analyze over time' is enabled in the sidebar, the Bliss score is computed at every measured timepoint and plotted as a synergy trend. This reveals whether synergy emerges, fades, or reverses over the course of treatment.

Per-timepoint p-values (Combination vs Drug A and vs Drug B, two-sample t-test) are included in the Synergy Metrics table. These are unadjusted for multiple testing across timepoints — interpret overall patterns rather than individual p-values at isolated days.