Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› cc-figures-tables
cc-figures-tables
GitHub用于构建或审核Cancer Cell稿件的多面板机制图表与表格。涵盖图像量化、面板逻辑、图像完整性及图例规范，确保代表图与重复实验数据匹配，符合期刊严格的出版标准。
Trigger Scenarios
展示缺乏量化的代表性图像（如WB、IHC）
多面板图表缺乏清晰叙事逻辑
担忧图像完整性问题（如拼接、过度处理）
构建图形摘要或补充显示项目
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill cc-figures-tables -g -y
SKILL.md
Frontmatter
{
"name": "cc-figures-tables",
"description": "Use when building or auditing multi-panel mechanistic figures and tables for a Cancer Cell (Cell Press) manuscript — quantification of images, panel logic, image integrity, and legends. It shapes display items; it does not choose statistical tests or write the abstract."
}
Figures & Tables (cc-figures-tables)
When to trigger
- Representative images (blots, IHC, micrographs) shown without quantification
- A multi-panel figure has no clear narrative logic
- Worried about image-integrity standards (splicing, over-processing)
- Building a graphical abstract or supplemental display items
Cancer Cell figure logic
Each main figure should advance one step of the mechanistic argument and be readable on its own:
- Panels flow left-to-right / top-to-bottom as a mini-narrative (perturbation → readout → mechanism → consequence).
- Mechanistic figures typically combine schematic + functional data + in vivo and/or human validation.
- A figure usually carries discovery, validation, and a control panel together.
- Keep one main message per figure; push orthogonal corroboration to supplement.
Quantify everything representative
The signature Cancer Cell requirement: a representative image must be paired with quantification across biological replicates.
| Representative image | Required quantification |
|---|---|
| Western blot | Densitometry across n independent blots (normalized to loading) |
| IHC / IF micrograph | Scored / counted across multiple fields and tumors (e.g., QuPath) |
| Flow plot | Summary bar/dot plot of % or MFI across replicates |
| Tumor photo | Tumor volume/weight curves with n, stats |
| Migration/invasion image | Quantified across fields/replicates |
| Single-cell UMAP | Proportions / DE stats, not just a colored embedding |
Show data points (superplots) for small n; never imply quantity from one picture.
Image integrity (Cell Press enforces)
- Keep uncropped, unprocessed raw blots/gels; include full-length blots in supplement with MW markers.
- No splicing of lanes without a clear dividing line and disclosure.
- Adjustments (brightness/contrast) must be linear and applied to the entire image, including controls.
- Do not duplicate, mirror, or reuse panels across figures.
- Disclose any grouping of images from different parts of the same gel.
Tables and the Key Resources Table
- The Key Resources Table (STAR Methods) is mandatory — see
cc-reporting-standards. - Data tables (cohorts, mutation lists, screen hits) go to supplement; keep main tables minimal.
- Define every abbreviation and unit in legends.
Legends
- State
n, what one replicate is, the statistical test, the error-bar definition, and scale bars. - Each panel letter referenced; magnification/scale bar on all micrographs.
- Antibodies / markers / treatments named in the legend.
Worked micro-example: fixing a "representative blot" panel
Before: Figure 2C is a single Western blot showing MARK7 down and cleaved-PARP up after treatment, captioned "representative of independent experiments."
Why a Cancer Cell reviewer rejects it as evidence: one image cannot support a quantitative claim; the
reader cannot see variance, n, or whether the effect is consistent.
After: Figure 2C keeps the representative blot but adds 2C′ — densitometry normalized to loading control across four independent lysates, plotted as a superplot (individual points + mean), with the test and exact p in the legend. The uncropped full-length blot with MW markers moves to Figure S2. Now the panel carries a claim, not an anecdote.
Panel-order sanity check (per main figure)
Read each figure left-to-right and ask whether it narrates one mechanistic step:
- Perturbation — what was changed (knockdown/knockout/inhibitor), with the validation of that perturbation.
- Readout — the phenotype or molecular consequence, quantified across replicates.
- Mechanism — the intermediate that connects perturbation to readout (binding, flux, signaling).
- Orthogonal confirmation — in vivo and/or human panel that the claim in the figure title needs.
If a panel does not serve the figure's one-sentence title, move it to the supplement.
Cancer Cell figure failure modes reviewers cite
- A migration/invasion or IHC figure with beautiful images but no per-field, per-tumor quantification.
- Tumor-growth curves without
nper arm, without the test, or truncated before humane endpoints. - Flow plots shown as gated dot-plots only, with no summary statistic across mice/donors.
- A graphical abstract that asserts an axis (arrows implying causality) the main figures never establish.
- Kaplan-Meier survival curves with no numbers-at-risk and no censoring shown.
- Contrast/brightness pushed non-linearly, or a loading-control band reused across panels.
Checklist
- Every figure advances one mechanistic step; panel order tells a story
- Each representative image is paired with replicate-level quantification
- Data points shown for small-n panels (superplots)
- Blots uncropped in supplement, MW markers visible, no undisclosed splicing
- Image adjustments linear and whole-image; no duplicated panels
- Scale bars on all micrographs; channels/markers labeled
- Legends give
n, test, error-bar definition - In vivo / human validation panels present where claims need them
- Graphical abstract (if used) conveys the mechanism at a glance
Anti-patterns
- One representative blot/image as the entire evidence (no quantification)
- Bar charts hiding tiny
n; undefined error bars - Spliced or over-contrasted blots; reused/duplicated panels
- Overcrowded figures mixing several unrelated messages
- Missing scale bars or unlabeled axes
- Beautiful schematic with no supporting data
Output format
【Per-figure message】Fig1: ... Fig2: ...
【Representative-without-quantification】flagged panels: [...]
【Image integrity】raw blots? linear adj? duplication check
【Legends】n / test / error bar / scale bar present? Y/N
【Validation panels】in vivo / human present where needed?
【Next step】cc-structured-abstract
Version History
- 1839142 Current 2026-07-05 12:26
