Polyomino

Polyomino reconstructs spatial transcriptomic profiles with single-cell resolution via a region-allocation method that optimizes data integration using gradient descent.

Key Features:

• Region-allocation method: Polyomino allocates cells to structured spatial regions that match the most significant biological information, inspired by the region-of-interest (ROI) concept from image processing.
• Gradient descent optimization: Polyomino uses gradient descent to optimize the integration of single-cell and spatial transcriptomic data.
• Scalability: Polyomino outperforms state-of-the-art methods by 10 to 1000 times in speed and is capable of integrating data sets containing millions of cells in a single run.
• Robustness to sequencing artifacts: Polyomino integrates data even in the presence of cell segmentation errors and imbalanced cell-type representations.

Scientific Applications:

• Gene expression pattern discovery: Polyomino uncovers originally hidden gene expression patterns in brain sections.
• Organogenesis research: Polyomino offers new insights into organogenesis through integrated spatial transcriptomic analysis.
• Tumor microenvironment analysis: Polyomino provides new insights into tumor microenvironments with exceptional efficiency and accuracy.

Methodology:

Polyomino uses gradient descent to allocate cells to structured spatial regions, optimizing the integration of single-cell and spatial transcriptomic data.