Skip to content

Match provenance (origin, non-exclusive matching); calibrated MS1 patch and cap-status fixes - #22

Merged
lfu46 merged 5 commits into
mainfrom
match-provenance
Oct 14, 2026
Merged

lfu46 merged 5 commits into
mainfrom
match-provenance

Conversation

@lfu46

@lfu46 lfu46 commented Oct 14, 2026

Copy link
Copy Markdown
Owner
  • canonical_match records each match's origin; match_peaks(exclusive=False) reports every identity a peak fits (defaults unchanged).
  • Calibrated MS1 patch: the smoothness check applies only to scans the cap cut.
  • ms1_cap_status: a cap needs at least 3 scans at the maximum peak count.

🤖 Generated with Claude Code

lfu46 and others added 5 commits October 14, 2026 10:38
…not cut

Nothing is interpolated there, and short end-of-run scans (80-150 peaks) can have a
held-out p99 just over 1 ppm, which refused NGlycoMM NG_12_R for no reason.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With few MS1 scans the single largest one already reached 5%, so an uncapped source
read as capped.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…allback

MatchedIon gains an origin field (default empty, so positional construction is
unchanged). Callers that keep evidence provenance can now tell a raw-fallback match
from a deisotoped one.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The default stays exclusive: a peak explains one ion, the first in priority order.
exclusive=False (also through canonical_match with the default matcher) lets each
ion take its nearest peak, so ambiguous peaks are visible as such.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@lfu46
lfu46 merged commit 1f1a769 into main Oct 14, 2026
2 checks passed
@lfu46
lfu46 deleted the match-provenance branch October 14, 2026 15:06
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