the two chimera measurements land where measurements live - #497
Merged
Merged
Conversation
…eel already ships Measurement for #407, under map #406. `chimera.split_name` plus `metadata.lookup_assembly` answer "is this name a chimera's, and of what?" with no read of the reference store and no network — but `Genome.components` says the record and never the table is the real test, and liulab-genome ships a test proving `ce11_ecHT115` seeded from somebody's FASTA is not a chimera. The offline predicate is about the NAME, it is capped at the seven rows the table ships, and the pin in this repo predates `chimera.py` entirely. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
…and not the aligner Research for #408 under map #406. STAR 2.7.11b read from source: a window is confined to one chromosome, so no proper pair spans two components; NH and MAPQ are both functions of nTrOutSAM, the chimera-wide reported-locus count; outFilterMultimapNmax is applied to that same count, so a read with 9 host and 2 contaminant loci is dropped at the default of 10. The load-bearing claim behind the map's standing decision 2 survives in substance and fails twice. STAR reports every locus within outFilterMultimapScoreRange (default 1) of the best, not only the best, so a read beating its cross-species hit by one point still carries NH:i:2 MAPQ:3. And map/star-umi passes --outSAMmultNmax 1, so its BAM holds one alignment record per template and "spans more than one component" is not observable in it at all. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Two research notes were written for closed wayfinder tickets and never reached
main— they havebeen sitting on their own branches since 2026-08-14, so the map that consumed them cites primary
sources nobody can read from a checkout.
docs/research/chimera-offline-detection.md— Can compose tell a chimera from its name alone, offline? #407: a chimeric name is decidable, and itscomponent list recoverable, from the name plus the wheel's shipped table, with no reference-store
read and no network; sound only while every component is a row in
assembly_metadata.tsv.docs/research/star-on-a-chimera.md— What STAR does with a chimeric reference, at split-contract resolution #408: a window carries one chromosome, so the split's blindspot is a flag (
--outSAMunmapped,--outFilterMultimapScoreRange) and not the aligner.Both commits are cherry-picked unchanged from
research/chimera-offline-detectionandresearch/star-on-a-chimera, authorship and message intact. Docs only — nosrc/, notests/.docs/research/**is in markdownlint'signoresand mkdocs'exclude_docs, so neither doc checkreaches these files; the gate is green.
Closes nothing — #406/#407/#408 are already closed. This lands the evidence they produced.
🤖 Generated with Claude Code