MIIT

MIIT integrates spatially resolved multi-omics data from Spatial Transcriptomics (ST) using Visium 10x ST and Mass Spectrometry Imaging (MSI) to align, map, and correlate molecular signals across serial tissue sections.

Key Features:

• Integration Workflow: Performs grid projection on spatial omics data to align and combine datasets from different imaging modalities.
• Registration Algorithm: Implements GreedyFHist, a registration algorithm based on the Greedy registration tool, for aligning stained serial sections and validated on 245 pairs of serial sections with improved registration performance over similar algorithms.
• Data Mapping: Maps MSI-pixels to the spot resolution of Visium 10x ST to enable spot-wise correlation between metabolite signals and mRNA expression or gene signature activities.
• Annotation Support: Supports pointset valued annotations, geojson data, and masks for diverse annotation types.

Scientific Applications:

• Prostate cancer spatial correlation analysis: Integration of ST and MSI on cancer-free tissue samples from seven prostate cancer patients to assess spot-wise correlations between gene signature activities related to citrate-spermine secretion (from ST) and metabolite levels such as citrate, spermine, and zinc (from MSI).

Methodology:

Computational steps explicitly include grid projection of spatial omics data, registration of stained serial sections using GreedyFHist (based on Greedy registration), mapping MSI-pixels to Visium 10x ST spot resolution, support for pointset/geojson/mask annotations, and validation of registration performance on a dataset of 245 pairs of serial sections.