A Physically-Realistic Simulator and Cosine-Based Decoder for MERFISH Spatial Transcriptomics
Bibliographic record
Abstract
Abstract Imaging-based spatial transcriptomics technologies have opened new avenues for studying cellular organization and gene expression within intact tissues. However, the accuracy of downstream analyses depends critically on the decoding step that reconstructs barcodes from fluorescence patterns and maps them to gene identities. Despite a growing number of decoding methods, systematic benchmarking has been limited. Here, we introduce Serval, a modular framework for developing and benchmarking decoding methods across diverse spatial transcriptomics platforms. Serval separates key decoding stages into independently configurable modules, enabling flexible integration of alternative algorithms. Using this framework, we develop the Cosine decoder, a novel method that improves transcript recovery by optimizing cosine similarity to known barcodes. We evaluate Cosine and baseline methods on synthetic and real MERFISH datasets, showing that Cosine achieves higher transcript recovery and superior correlation with expression references compared to existing methods. Furthermore, we demonstrate that the Serval framework generalizes beyond MERFISH. By extending to the DART-FISH platform, we show that Cosine improves transcript recovery, clustering stability and supports more direct annotation of complex biological structures such as the human primary motor cortex. These results establish that modular decoding frameworks facilitate robust, platformagnostic benchmarking, ultimately supporting more accurate spatial transcriptomics analysis across diverse biological samples.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| 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 | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".