Securing diagonal integration of multimodal single-cell data against ambiguous mapping
Bibliographic record
Abstract
MOTIVATION: Recent advances in single-cell multimodal omics technologies enable the exploration of cellular systems at unprecedented resolution, leading to the rapid generation of multimodal datasets that require sophisticated integration methods. Diagonal integration has emerged as a flexible solution for integrating heterogeneous single-cell data without relying on shared cells or features. However, the absence of anchoring elements introduces the risk of artificial integrations, where cells across modalities are incorrectly aligned due to ambiguous mapping. RESULTS: To address this challenge, we propose SONATA (Securing diagOnal iNtegrATion against Ambiguous) mapping, a novel diagnostic method designed to detect potential artificial integrations resulting from ambiguous mappings in diagonal data integration. SONATA identifies ambiguous alignments by quantifying cell-cell ambiguity within the data manifold, ensuring that biologically meaningful integrations are distinguished from spurious ones. It is worth noting that SONATA is not designed to replace any existing pipelines for diagonal data integration; instead, SONATA works simply as an add-on to an existing pipeline for achieving more reliable integration. Through a comprehensive evaluation on both simulated and real multimodal single-cell datasets, we observe that artificial integrations in diagonal data integration are widespread yet surprisingly overlooked, occurring across all mainstream diagonal integration methods. We demonstrate SONATA's ability to safeguard against misleading integrations and provide actionable insights into potential integration failures across mainstream methods. Our approach offers a robust framework for ensuring the reliability and interpretability of multimodal single-cell data integration. AVAILABILITY AND IMPLEMENTATION: The source code is available at (https://github.com/batmen-lab/SONATA).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".