GenOT: generative optimal transport enables spatiotemporal interpolation and generation in cross-platform spatial...

Area of Science:

• Computational Biology
• Genomics
• Bioinformatics

Background:

• Spatial transcriptomics enables gene expression analysis within tissue context.
• Integrating and interpolating data across heterogeneous samples and platforms remains a significant challenge.

Purpose of the Study:

• To develop a novel generative framework, GenOT, for efficient spatiotemporal interpolation of spatial transcriptomics data.
• To address challenges in cross-slice and cross-platform data integration and reconstruction of gene expression dynamics.

Main Methods:

• GenOT combines multi-scale graph self-supervised contrastive learning with optimal transport barycenter theory.
• An optimal transport barycenter-based interpolation algorithm models spatial distribution differences for reconstruction.

Main Results:

• GenOT demonstrates superior performance in spatial domain identification compared to existing methods.
• The framework achieves state-of-the-art results in cross-platform interpolation.
• GenOT effectively reconstructs developmental trajectories from spatial transcriptomics data.

Conclusions:

• GenOT provides an efficient and robust solution for integrating and interpolating spatial transcriptomics data.
• The proposed method advances the analysis of spatiotemporal gene expression dynamics in heterogeneous biological samples.