Source Nature · Published · Open the original ↗
Capability
Visual LLM-guided consensus spatial domain detection with L-STAR
Zhao and Ji present L-STAR, a framework that uses a large language model to identify and combine high-performing methods for mapping patterns of gene activity within tissues, improving the accuracy, robustness and interpretability of the analysis.
Spatial transcriptomics measures gene activity while preserving the locations of cells within tissue. A key analysis step is to divide the tissue into spatial domains, but the performance of computational methods varies considerably across datasets, and no single method consistently performs best. Here we show that a large language model can compare spatial domain maps and identify the better-performing methods. We developed L-STAR, which visually compares and ranks candidate methods, then builds a consensus map from the highest-ranked results. […]
The opening of the report, quoted unchanged from Nature. The full text continues at the source.