cc-workflow

Cancer Cell Workflow (cc-workflow)

Overview

This is the router. It does not replace any specialized skill — it tells you which cc- skill to use at your current stage* of a Cancer Cell (Cell Press) molecular / translational oncology manuscript.

Default assumption: unless the user says otherwise, treat the target as Cancer Cell, where the bar is a clear molecular mechanism validated across orthogonal systems (cells + in vivo + ideally human data), reported with STAR Methods rigor, framed for translational relevance without overclaiming.

When to trigger

• The user asks "what should I do next?"
• A draft arrives and you must diagnose the current bottleneck
• Work is thrashing between experiments, figures, and writing
• A decision letter / reviewer reports arrive and you must switch into revision mode

Routing table

Default order

1. cc-scope-fit — confirm mechanism + translational relevance before investing more
2. cc-study-design — design / audit orthogonal validation, controls, replicates, in vivo rigor
3. cc-reporting-standards — STAR Methods, Key Resources Table, authentication, RRIDs
4. cc-statistics — define n, pick tests, correct for multiplicity, label error bars
5. cc-figures-tables — multi-panel mechanistic figures with quantification + image integrity
6. cc-structured-abstract — Summary, Highlights, eTOC blurb, graphical abstract
7. cc-ethics-registration — approvals, consent, biosafety, data-deposition statements
8. cc-writing-style — Cell Press prose and claim calibration (polish stage)
9. cc-cover-letter — significance / fit / suggested reviewers
10. cc-submission — pre-submission preflight
11. cc-peer-review-revision — after reviewer reports

cc-writing-style and cc-structured-abstract are late-stage polish — do not finalize them while the mechanism or in vivo validation is still missing.

Decision shortcuts

• "It's only in cell lines" → cc-study-design (add in vivo / human validation)
• "Reviewers will ask if the cell line is authenticated" → cc-reporting-standards
• "I used n=3 wells from one experiment" → cc-statistics (pseudo-replication)
• "I have a beautiful blot but no densitometry" → cc-figures-tables
• "My title promises a therapy but I have no in vivo efficacy" → cc-scope-fit then cc-writing-style
• "No GEO accession yet" → cc-ethics-registration / cc-submission
• "Three reviewers, consultative cross-review" → cc-peer-review-revision

Differences vs. JAMA-style clinical packs

If the work is a clinical trial or epidemiological cohort with patient-level outcomes as the core unit, a clinical-trial pack (CONSORT / STROBE / registration) fits better. Cancer Cell's unit is a mechanism validated across systems, not a trial endpoint.

Worked routing example

"We have RNA-seq showing MARK7 correlates with CAF activation in a patient cohort, plus a knockdown migration phenotype in one PDAC line. We want to submit to Cancer Cell."

Route it:

1. cc-scope-fit — a correlation + single-line phenotype is off-fit on both pillars; the mechanism is not established and there is no in vivo/orthogonal validation. Gate here first.
2. cc-study-design — plan the missing spine: in vivo perturbation in an immunocompetent model, a second cell system, and a mechanistic intermediate linking MARK7 to the phenotype.
3. Only once that evidence exists do cc-reporting-standards → cc-statistics → cc-figures-tables apply; drafting front matter (cc-structured-abstract, cc-writing-style) before then is premature.

The router's job is to stop a promising-but-thin story from being polished into a confident desk reject.

Stage diagnosis cues

Evidence-spine manifest

Cancer Cell's unit is a mechanism validated across orthogonal systems. Track the spine as a manifest and route on the first MISSING: front-matter polish is premature until cells + in vivo + (ideally) human evidence converge on one mechanism.

evidence_spine:
  mechanism:        OK | UNCLEAR      # molecular intermediate linking cause -> phenotype
  cell_system_1:    OK | MISSING
  cell_system_2:    OK | MISSING      # orthogonal line / method, not a technical replicate
  in_vivo:          OK | MISSING      # perturbation in an immunocompetent model
  human_data:       OK | MISSING      # patient cohort / clinical specimens
  rigor:
    key_resources_table: yes | no     # cc-reporting-standards
    cell_line_authenticated: yes | no
    n_defined_no_pseudorep: yes | no  # cc-statistics
    image_quantification: yes | no    # cc-figures-tables
  deposition:
    geo_or_pride_accession: yes | no  # cc-ethics-registration
route_rule: "first MISSING/UNCLEAR above -> its listed skill; do not draft Summary/Highlights until the spine is OK"

Anti-patterns

• Do not skip cc-scope-fit — editors triage on mechanism + translational fit first
• Do not let cc-figures-tables polish panels before cc-statistics has fixed n and tests
• Do not let cc-peer-review-revision draft a response before the revised experiments / text exist
• Do not route to front-matter polish while the in vivo or human-validation spine is still missing
• Do not treat a presubmission inquiry as a substitute for the scope gate — run cc-scope-fit regardless