Skip to content

Advance countMatMult stride once per gene, not once per dedup option (fixes #2704) - #2705

Open
danielsharritt wants to merge 1 commit into
alexdobin:masterfrom
danielsharritt:fix-multimapper-dedup-stride
Open

danielsharritt wants to merge 1 commit into
alexdobin:masterfrom
danielsharritt:fix-multimapper-dedup-stride

Conversation

@danielsharritt

Copy link
Copy Markdown

Moves SoloFeature_collapseUMIall.cpp:530 out of the indDedup loop so each multimapping gene occupies one countMatMult group, which is the layout ParametersSolo.cpp:661-665 and SoloFeature_outputResults.cpp:151-197 already assume. With more than one --soloUMIdedup option the current code writes N records per multimapping gene to every UniqueAndMult-*.mtx, N-1 of them on feature index 1. Tested on pbmc_1k_v3: with this change the three-option run's UniqueAndMult-EM_umiDedup-1MM_Directional_UMItools.mtx is byte-identical to the single-option UniqueAndMult-EM.mtx (md5 6c89f52b389b2432b0c5571bfbf8dcf9). Full evidence, logs and reproduction in #2704. Fixes #2704.

🤖 Generated with Claude Code

…up option

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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.

Multiple --soloUMIdedup options break UniqueAndMult-*.mtx: extra records on feature index 1

1 participant