Repository navigation
specs/030: hierarchical root-complex decomposition (LNG #97) + catalog.sh guard fix - #75
Merged
Merged
Conversation
#97) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…hing leaked (grep exit 1 under set -euo pipefail) Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…t +228 (238/10, p=9e-58), IFN a/b +200 and +28 outside it, experimental +2 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…of 222 severed routes repaired outside IFN/DSB); pre-register re-measure on hier2 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…33), +22 outside IFN, exp 0; TP53 -81 traced to uuid order, IFN -25 to the root-copy preference Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
… read 186 vs 166 'edge missing' on identical input under different hash seeds); route_breaks --steps writes the broken step per case Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…n represent (into root complexes and their nesting); route_breaks labels each severed route faithful vs composition-only and --faithful walks the representable route Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…reak at complexes Reactome never forms (curation-review list), 80 entity->reaction are all copy multiplicity (sink-bridge family, measured closed), ~31 cases for an any-copy pool Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…erarchy) -- held-out 87.60%/0.8388, all-pathways 85.18%/0.8188 (TP53 uuid draw is +131 of it), experimental 67.46%; CLAUDE.md indexes specs/030, 031 Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
…tor decomposes (roots used by a reaction; produced nested complexes are joins); route_breaks default output unchanged; catalog.sh guard reads names via compgen -e, with a runtime test red on both prior versions; RESULTS box from the bench's own holdout.txt (in-release keeps unperturbable cases: held-out 87.24%/0.8346), like-for-like curator lead +1.04pp, TP53 draw disclosed on both axes; specs/030 TP53 counts and table row; specs/031 re-split on the corrected oracle, every class by script Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
adamjohnwright
marked this pull request as ready for review
September 26, 2026 17:18
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
scripts/catalog.sh: theLNG_*leak guard killed every build silently when nothing leaked, because grep exits 1 underset -euo pipefail. main has the broken version.Still to add before merge: the canonical rebuild on the merged generator, the re-baseline, and the
docs/RESULTS.md/ CLAUDE.md updates.🤖 Generated with Claude Code