Spatial-linked alignment tool (SLAT) for aligning heterogenous slices properly

Abstract

Spatially resolved omics technologies reveal the spatial organization of cells in various biological systems. Integrative and comparative analyses of spatial omics data depend on proper slice alignment, which should take both omics profiles and spatial context into account. Here we propose SLAT (Spatially-Linked Alignment Tool), a graph-based algorithm for efficient and effective alignment of spatial omics data. Adopting a graph adversarial matching strategy, SLAT is the first algorithm capable of aligning heterogenous spatial data across distinct technologies and modalities. Systematic benchmarks demonstrate SLAT's superior precision, robustness, and speed compared to existing methods. Applications to multiple real-world datasets further show SLAT's utility in enhancing cell-typing resolution, integrating multiple modalities for regulatory inference, and mapping fine-scale spatial-temporal changes during development. The full SLAT package is available at https://github.com/gao-lab/SLAT.

DeST-OT: Alignment of Spatiotemporal Transcriptomics Data

“…al. [43] used Stereo-seq [5] to measure gene expression in the axolotl telencephalon at five development timepoints: three embryonic stages (44,54,57), Juvenile stage, and Adult stage (Fig. 3a).…”

“…STalign [8] is an image registration method finding a diffeomorphism between the H&E images of two spatial slices. GPSA [18] uses Gaussian processes to register spatial slices onto a common coordinate system, while SLAT [44] relies on graph neural networks and adversarial learning.…”

“…moscot allows for cell growth and death using a curated set of proliferation and apoptosis genes as prior knowledge but optimizes an objective function that encourages static shape-matching. Other works [18,44] do not necessarily quantify growth, and often have methodological and robustness limitations.…”

“…SLAT [44] We use scSLAT.model.run_SLAT of the scSLAT package for computing an embedding for each spot in each timepoint. We then use scSLAT.model.spatial_match to compute an alignment between spots across the two timpoints.…”

ST-GEARS: Advancing 3D Downstream Research through Accurate Spatial Information Recovery

“…Moreover, its hypothesis space involves readout prediction in addition to coordinates alignment, causing uncertainty in direction of gradient descent, and vulnerabilities to input noises. Another alignment approach, Spatial-linked alignment tool (SLAT)[30] also focuses on anchors construction between sections, yet it doesn’t provide a methodology to construct 3D transcriptomics profile. Other tools focus on analysis and visualization of 3d data, such as Spateo[31], VT3D[32] and StereoPy[33].…”