Skip to content

Describe sbs96_compare.py as it is: stdlib-only, kept untested - #820

Merged
joshfactorial merged 1 commit into
developfrom
docs/sbs96_compare_description
Oct 6, 2026
Merged

joshfactorial merged 1 commit into
developfrom
docs/sbs96_compare_description

Conversation

@joshfactorial

Copy link
Copy Markdown
Collaborator

CLAUDE.md and the eidolon/src/compare_af/mod.rs header both said scripts/delta/sbs96_compare.py parses SigProfiler's output, and CLAUDE.md gave that as the one justification for keeping it in Python. The script uses only the Python standard library: it reads a FASTA and two VCFs and computes the SBS-96 cosine itself.

Both descriptions now say what it does, that it produced the HCC1395 0.72 → 0.99 figure (#372), and that it has no current caller. Decision (2026-10-06): keep it as is, untested. Tests get added if it is needed again; otherwise it goes when the validation scripts are archived.

Comment and docs only. No behavior change, no version impact.

Covers item 6 of #819.

🤖 Generated with Claude Code

CLAUDE.md and the compare_af header said it parses SigProfiler output,
which was the stated reason it stays Python. It reads a FASTA and two
VCFs and needs no external tool. It produced the #372 HCC1395 cosine
figure and has no current caller; it stays as is, with tests added only
if it is needed again.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@joshfactorial
joshfactorial merged commit c7f3d85 into develop Oct 6, 2026
7 checks passed
@joshfactorial
joshfactorial deleted the docs/sbs96_compare_description branch October 6, 2026 12:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant