Experimental Design (cc-study-design)
When to trigger
- The mechanism rests on a single system (e.g., cell lines only)
- Reviewers will ask "is this true in vivo?" or "does it hold in patients?"
- Controls, replicates, or sample-size logic are unclear
- Planning mouse / PDX / organoid experiments and unsure about rigor elements
Orthogonal validation ladder
Cancer Cell expects a mechanism corroborated across independent, complementary systems. Build the strongest ladder the biology allows:
| Layer |
Systems |
Role |
| In vitro |
Cell lines (multiple, authenticated), primary cells, co-cultures, biochemistry |
Establish mechanism, gain/loss-of-function, epistasis |
| Functional genetics |
CRISPR KO/KI, shRNA/siRNA with rescue, point mutants, degrons |
Causality and specificity |
| In vivo |
GEMM, syngeneic, xenograft, PDX, orthotopic, metastasis models |
Mechanism operates in a tumor in an organism |
| 3D / ex vivo |
Tumor organoids, patient-derived organoids, slice cultures, spheroids |
Bridge to human, drug response |
| Human |
Patient tumor samples, TMAs, scRNA-seq, public cohorts (TCGA), outcome data |
Translational anchor / clinical relevance |
A Cancer Cell paper typically spans in vitro + in vivo + a human anchor. Decide early which layers carry the causal claim and which provide corroboration.
Controls (the parts reviewers attack)
- Genetic perturbation: include rescue / add-back; use ≥2 independent sgRNAs or shRNAs (or rule out off-target); non-targeting / scramble control matched to delivery.
- Pharmacology: vehicle control, dose-response, on-target genetic phenocopy, and where possible a resistant mutant or analog.
- In vivo: isotype/vehicle controls, littermate or co-housed controls for GEMMs, tumor-volume and endpoint pre-defined.
- Antibody / staining: isotype, KO/KD-validated, single-stain compensation for flow.
Replicates: biological vs technical
- Biological replicates = independent biological units (separate mice, independent cell passages/cultures, distinct patients). These define
n.
- Technical replicates (duplicate wells, repeat measurements) reduce measurement noise but do not increase
n.
- Report both clearly; never inflate
n with technical replicates (see cc-statistics).
Sample size, randomization, blinding (especially animals)
- Power / sample size: justify mouse
n (effect size + variance from pilot or literature); state the basis even if informal.
- Randomization: allocate animals to arms randomly (e.g., when tumors reach a set volume), not by cage convenience.
- Blinding: blind outcome assessment (tumor measurement, histology scoring, imaging quantification) wherever feasible.
- Inclusion/exclusion: pre-define humane endpoints and exclusion criteria; report all animals/samples and any exclusions.
Human-sample design
- Define cohort, inclusion/exclusion, and how samples link to outcomes.
- Power survival / association analyses; pre-specify primary comparison.
- Note IRB/consent (route to
cc-ethics-registration).
Checklist
Anti-patterns
- Conclusions from a single cell line or a single system
- shRNA/CRISPR phenotype with no rescue (off-target not excluded)
- "n=3" that is three wells of one experiment (pseudo-replication)
- Mouse experiments with no randomization, no blinding, no power basis
- Therapeutic efficacy claimed without an in vivo tumor model
- Selecting samples post hoc to fit the hypothesis
Output format
【Causal claim layer】in vitro / in vivo / human
【Orthogonal systems planned】...
【Controls per perturbation】rescue + 2 reagents? vehicle/isotype?
【Replicates】biological n = ... ; technical handled separately
【Animal rigor】power basis / randomization / blinding / endpoints
【Human anchor】cohort + primary comparison
【Gaps to close before submission】...
【Next step】cc-reporting-standards (rigor reporting) or cc-statistics
Source: brycewang-stanford/Awesome-Journal-Skills → Cancer-Cell-Skills/skills/cc-study-design/SKILL.md
1---2name: cc-study-design3description: Use when designing or auditing the experimental plan for a Cancer Cell (Cell Press) study — orthogonal validation across in vitro, in vivo, and human tumor systems, with controls, replicates, power, randomization, and blinding. It plans design; it does not run statistics or write figures.4---567# Experimental Design (cc-study-design)89## When to trigger1011- The mechanism rests on a single system (e.g., cell lines only)12- Reviewers will ask "is this true in vivo?" or "does it hold in patients?"13- Controls, replicates, or sample-size logic are unclear14- Planning mouse / PDX / organoid experiments and unsure about rigor elements1516## Orthogonal validation ladder1718Cancer Cell expects a mechanism corroborated across **independent, complementary systems**. Build the strongest ladder the biology allows:1920| Layer | Systems | Role |21|-------|---------|------|22| In vitro | Cell lines (multiple, authenticated), primary cells, co-cultures, biochemistry | Establish mechanism, gain/loss-of-function, epistasis |23| Functional genetics | CRISPR KO/KI, shRNA/siRNA with rescue, point mutants, degrons | Causality and specificity |24| In vivo | GEMM, syngeneic, xenograft, **PDX**, orthotopic, metastasis models | Mechanism operates in a tumor in an organism |25| 3D / ex vivo | Tumor **organoids**, patient-derived organoids, slice cultures, spheroids | Bridge to human, drug response |26| Human | Patient tumor samples, TMAs, scRNA-seq, public cohorts (TCGA), outcome data | Translational anchor / clinical relevance |2728A Cancer Cell paper typically spans in vitro + in vivo + a human anchor. Decide early which layers carry the **causal** claim and which provide corroboration.2930## Controls (the parts reviewers attack)3132- **Genetic perturbation:** include rescue / add-back; use ≥2 independent sgRNAs or shRNAs (or rule out off-target); non-targeting / scramble control matched to delivery.33- **Pharmacology:** vehicle control, dose-response, on-target genetic phenocopy, and where possible a resistant mutant or analog.34- **In vivo:** isotype/vehicle controls, littermate or co-housed controls for GEMMs, tumor-volume and endpoint pre-defined.35- **Antibody / staining:** isotype, KO/KD-validated, single-stain compensation for flow.3637## Replicates: biological vs technical3839- **Biological replicates** = independent biological units (separate mice, independent cell passages/cultures, distinct patients). These define `n`.40- **Technical replicates** (duplicate wells, repeat measurements) reduce measurement noise but **do not** increase `n`.41- Report both clearly; never inflate `n` with technical replicates (see `cc-statistics`).4243## Sample size, randomization, blinding (especially animals)4445- **Power / sample size:** justify mouse `n` (effect size + variance from pilot or literature); state the basis even if informal.46- **Randomization:** allocate animals to arms randomly (e.g., when tumors reach a set volume), not by cage convenience.47- **Blinding:** blind outcome assessment (tumor measurement, histology scoring, imaging quantification) wherever feasible.48- **Inclusion/exclusion:** pre-define humane endpoints and exclusion criteria; report all animals/samples and any exclusions.4950## Human-sample design5152- Define cohort, inclusion/exclusion, and how samples link to outcomes.53- Power survival / association analyses; pre-specify primary comparison.54- Note IRB/consent (route to `cc-ethics-registration`).5556## Checklist5758- [ ] Mechanism validated across ≥2 orthogonal systems; causal layer identified59- [ ] In vivo evidence present (GEMM / xenograft / PDX / orthotopic) where claims require it60- [ ] Genetic perturbations include rescue and ≥2 independent reagents61- [ ] Matched controls defined for every perturbation and treatment62- [ ] Biological vs technical replicates distinguished; `n` = biological units63- [ ] Animal sample size justified; randomization and blinding specified64- [ ] Inclusion/exclusion criteria and humane endpoints pre-defined65- [ ] Human-sample cohort and primary comparison pre-specified6667## Anti-patterns6869- Conclusions from a single cell line or a single system70- shRNA/CRISPR phenotype with no rescue (off-target not excluded)71- "n=3" that is three wells of one experiment (pseudo-replication)72- Mouse experiments with no randomization, no blinding, no power basis73- Therapeutic efficacy claimed without an in vivo tumor model74- Selecting samples post hoc to fit the hypothesis7576## Output format7778```79【Causal claim layer】in vitro / in vivo / human80【Orthogonal systems planned】...81【Controls per perturbation】rescue + 2 reagents? vehicle/isotype?82【Replicates】biological n = ... ; technical handled separately83【Animal rigor】power basis / randomization / blinding / endpoints84【Human anchor】cohort + primary comparison85【Gaps to close before submission】...86【Next step】cc-reporting-standards (rigor reporting) or cc-statistics87```8889---9091**Source:** [`brycewang-stanford/Awesome-Journal-Skills`](https://github.com/brycewang-stanford/Awesome-Journal-Skills) → `Cancer-Cell-Skills/skills/cc-study-design/SKILL.md`
thedixitjain