Free AI-Assisted

Power & Effect Navigator

Calculate sample sizes, power curves, and reviewer-ready IACUC grant justification text for standard tests and RNA-seq—free in browser. Built-in AI agent assistant support.

Open tool workspace

Need help with bioinfo analysis? See our bioinformatics services.

Key facts

Key facts about Power & Effect Navigator
FactValue
Test typesTwo-sample/paired/one-sample t-test, one-way ANOVA (2–20 groups), two proportions, RNA-seq (negative-binomial)
Target solverSample size N (analyzable & total enrolled), statistical power (1−β), or minimum detectable effect (MDE / fold-change)
RNA-seq controlsFDR threshold (0–0.5), total genes, expected DE fraction, biological dispersion, mean read count
Cohort adjustmentsAllocation ratio (N₂/N₁), expected dropout rate (0–90%), pilot data uncertainty shrinkage (N_pilot ≥ 4)
VisualizationsSVG power vs. sample size curve with optional ±25% effect size sensitivity overlay lines
ExportsCopyable results, formatted text (IACUC, NIH grant, manuscript methods), power curve CSV and SVG/PNG
Runs in browserYes — fully client-side calculation, no server data upload
Account requiredNo
AI assistantBuilt-in; assists with effect size selection, parameter validation, and interpreting power curves

What it does

Biomedical researchers often struggle to justify study cohort sizes, leaving grant applications and IACUC protocols vulnerable to rejection over arbitrary sample estimates. Power & Effect Navigator removes the friction of prospective sample size planning by unifying classical biostatistical calculations, RNA-seq differential expression power modeling, animal attrition adjustments, and automated narrative justification writing into a single client-side web application—eliminating manual formula errors, software licensing constraints, and vague reviewer justifications.

Researchers begin by selecting their study design—two-sample, paired, or one-sample t-test, one-way ANOVA across 2 to 20 groups, two-proportion z-tests, or negative-binomial RNA-seq. Users toggle their target solver to calculate required sample size N, achieved statistical power (1−β), or minimum detectable effect (MDE or fold-change). Inputs include significance level α (or FDR target for RNA-seq), allocation ratio N₂/N₁, expected attrition rate, and effect size—entered directly, chosen via Cohen benchmarks, or adjusted for pilot noise using N_pilot shrinkage.

Upon calculation, the tool instantly displays analyzable sample size alongside total enrolled cohort counts required to absorb planned animal dropout. The right-hand panel renders an interactive SVG power curve featuring optional ±25% sensitivity comparison lines and a collapsible formula drawer showing non-centrality parameters. Switching to the Justification Writer tab transforms raw statistical outputs into customizable narrative paragraphs formatted specifically for IACUC animal protocols, NIH grant proposals, or manuscript methods sections.

Why researchers use it

  • Avoid underpowered animal study designs that trigger IACUC protocol rejections
  • Generate reviewer-ready sample size justification paragraphs for NIH grant applications
  • Plan RNA-seq cohorts using FDR targets and biological dispersion parameters
  • Inflate enrolled sample sizes automatically to compensate for expected animal attrition
  • Prevent undersized main studies by applying conservative shrinkage to pilot data
  • Export publication-ready power curve graphics and raw CSV data tables

Best for

  • Drafting IACUC animal care and use protocols requiring formal sample size justification
  • Writing NIH, DOD, or foundation grant applications with explicit power calculations
  • Budgeting RNA-seq differential gene expression experiments with targeted FDR control
  • Sizing two-arm or multi-group preclinical drug treatment vs control experiments
  • Determining minimum detectable fold-change when sample availability is strictly constrained

When to use this vs alternatives

Choose Power & Effect Navigator when prospective study planning requires both statistical computation and reviewer-ready justification text in one browser session. Use Biostat Test Navigator when your experiment is complete and you need to select the correct hypothesis test for collected data, or ΔΔCt Clarity for qPCR fold-change analysis. Desktop G*Power or R packages (pwr, RnaSeqSampleSize) remain suitable for custom multi-level linear mixed models or automated batch R pipelines.

What makes it different

Unlike legacy desktop tools and single-purpose calculators that return isolated numbers, Power & Effect Navigator bridges classical biostatistics and omics power in one interface, automatically adjusts for animal dropout and pilot uncertainty, and generates ready-to-copy protocol justification paragraphs without requiring R coding or local software installations.

Biomedical researchers switch from desktop software and static spreadsheets to Power & Effect Navigator when grant deadlines demand clear narrative rationale rather than raw calculations. The tool eliminates transcription errors between calculators and grant documents while preventing underpowered studies caused by uncorrected pilot effect sizes or omitted animal attrition buffers.

How to get started

  1. Select your experimental design tab (t-test, ANOVA, Proportions, or RNA-seq) at the top of the Calculator tab.
  2. Choose your target variable under Solve for: Sample size N, Power (1−β), or Minimum Detectable Effect size (MDE).
  3. Enter required design parameters including significance level α (or FDR), desired power, expected dropout rate, and effect size or fold-change.
  4. Click Calculate to review analyzable N, total enrolled N, SVG power curve, and the collapsible Formula & Assumptions drawer.
  5. Open the Justification Writer tab, select your preferred template (IACUC, NIH Grant, or Manuscript), edit the generated text, and click Copy to clipboard.

Frequently asked questions

Does Power & Effect Navigator perform post-hoc power analysis on completed experiments?
No. Post-hoc power analysis computed from observed p-values is mathematically tautological and widely discouraged by biostatisticians because it provides no information beyond the p-value itself. Power & Effect Navigator supports only prospective (a priori) power calculations to assist researchers in designing prospective experiments and drafting valid grant or IACUC protocol justifications.
How does the tool calculate sample size for RNA-seq differential expression studies?
RNA-seq mode utilizes a negative-binomial Wald test model that incorporates biological dispersion, mean read counts, total gene count, and expected proportion of differentially expressed genes. The individual gene significance level is automatically adjusted for multi-testing using the False Discovery Rate threshold (α_adj = FDR × DE_fraction), delivering accurate cohort estimates without running command-line bioconductor packages.
What is pilot data uncertainty correction and when should I turn it on?
Small pilot studies frequently overestimate true effect sizes due to sampling noise, leading researchers to design undersized main experiments. When you enable Pilot Uncertainty Correction and input your pilot sample size N_pilot, the tool applies statistical shrinkage (d_adj = d_pilot × √((N_pilot−3)/(N_pilot−1))) to produce a realistic, conservative effect size for study planning.
How does the tool account for animal dropout and attrition?
When you enter an expected dropout rate (e.g., 10% for animal post-operative mortality or tissue loss), the calculator computes both analyzable sample size (N_analyzable) and total enrolled cohort size (N_enrolled = ⌈N_analyzable / (1 − dropout)⌉). This ensures your protocol requests sufficient animals to maintain statistical power after anticipated attrition.
Can I use an AI agent or MCP server with Power & Effect Navigator?
Yes. Pepkio Tools features a built-in workspace assistant that can validate design parameters, suggest Cohen effect sizes, and explain power curve sensitivity. External AI agents (such as Cursor or Claude Desktop) can also trigger calculations programmatically over [API & MCP](/tools/developers) using the pepkio_power-and-effect-navigator tool identifier.

Client source code & registry

Open tool workspaceBack to tools

Last updated . Pepkio builds free lab calculators alongside bioinformatics CRO services.