HiCat: a semi-supervised approach for cell type annotation
Bibliographic record
Abstract
Existing cell type annotation methods face significant hurdles: supervised approaches often fail to differentiate between novel cell types not present in reference data, while unsupervised techniques can suffer from cluster impurity and difficulties in robustly distinguishing multiple distinct unknown cell populations. This critical gap motivated the development of HiCat, a semi-supervised pipeline specifically designed to overcome these limitations. HiCat is a semi-supervised pipeline that integrates both approaches, leveraging reference (labeled) and query (unlabeled) genomic data to simultaneously enhance annotation accuracy for known cell types and improve the discovery and differentiation of novel ones. HiCat follows a structured pipeline: (1) removing batch effects and generate a low-dimensional embedding; (2) nonlinear dimensionality reduction for capturing key patterns; (3) unsupervised clustering for proposing novel cell type candidates; (4) merging multi-resolution features from previous steps into a condensed feature space; (5) training a classifier on reference data for supervised annotation; and (6) resolving inconsistencies between supervised predictions and unsupervised clusters to finalize annotations, particularly for unseen types. Performance was evaluated across 10 public genomic datasets and perform a case study on a molecular cell atlas of the human lung. HiCat demonstrated superior performance in both known cell type classification and novel cell type identification. In benchmark evaluations, HiCat consistently outperformed existing methods, critically excelling in identifying and distinguishing multiple novel cell types. HiCat presents a robust framework for scRNA-seq cell annotation, improving classification accuracy and novel type identification. In addition, it provides a scalable and transferable solution for biomedical research, directly addressing key challenges in automated cell annotation.
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".