STIFT: spatiotemporal transcriptomics integration through spatially informed multi-timepoint bridging

Recent advances in spatial transcriptomics have highlighted the need for integrating spatiotemporal transcriptomics data, defined as spatially resolved gene expression profiles captured across sequential time points in developmental or regenerative processes. We present STIFT (SpatioTemporal Integration Framework for Transcriptomics) specifically designed for integrating spatiotemporal transcriptomics data. STIFT is a three-component framework combining developmental spatiotemporal optimal transport, spatiotemporal graph construction, and a graph attention autoencoder informed by temporal triplet learning. STIFT integrates large-scale 2D or 3D spatiotemporal transcriptomics data, enabling batch effect removal, spatial domain identification, trajectory inference and exploration of developmental dynamics. Applied to axolotl brain regeneration, mouse embryonic development, and 3D planarian regeneration datasets, STIFT removes batch effects and achieves clear spatial domain identification while preserving temporal developmental patterns and biological variations across hundreds of thousands of spots, demonstrating its effectiveness and specificity in integrating spatiotemporal transcriptomics data.