Single cell clustering method and system based on multi-granularity Transform

By integrating genomic association information through multi-granularity Transformer, the problem of high-dimensional sparse noise interference in single-cell clustering is solved, and more accurate cell feature representation and clustering are achieved.

CN121963901APending Publication Date: 2026-05-01SHANXI UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202610027196.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-09
Publication Date
2026-05-01

AI Technical Summary

Technical Problem

Existing single-cell clustering methods struggle to effectively integrate genomic association information when processing high-dimensional, sparse, and noisy scRNA-seq data, resulting in limited clustering accuracy.

Method used

We employ a multi-granularity Transformer-based approach, which constructs a multi-level associated cell representation through multi-granularity feature extraction, dual encoder training, and hybrid loss optimization. This captures the global functional modules and local collaborative expression patterns of the genome, thereby enhancing feature discriminability.

Benefits of technology

It effectively alleviates interference from high-dimensional sparse noise, significantly improves the accuracy and robustness of single-cell clustering, and enhances the discriminative power and biological significance of feature representation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention belongs to the technical field of single cell clustering, and particularly relates to a single cell clustering method and system based on multi-granularity Transform. In order to solve the problems of insufficient cell genome relation extraction, limited clustering accuracy and the like in scRNA-seq data, the method provides a multi-granularity feature learning framework, hierarchical Token division is carried out on gene expression data, multi-level association from local gene collaboration to a global function module is captured by utilizing an attention mechanism, and the clustering accuracy is improved. And thus, cell global representation with higher discrimination is constructed. Furthermore, by introducing a cluster contrast learning mechanism, feature collision possibly caused by negative sample pairs in traditional contrast learning is avoided, and the clustering friendliness of a feature space is enhanced.
Need to check novelty before this filing date? Find Prior Art