IAMSAM : Image-based Analysis of Molecular signatures using the Segment-Anything Model

IAMSAM (Image-based Analysis of Molecular signatures using the Segment-Anything Model) is a user-friendly web-based tool designed to analyze ST data. This repository contains the code and resources to utilize the functionalities of IAMSAM described in our paper.

IAMSAM utilizes the Segment-Anything Model for H&E image segmentation, which allows for morphological guidance in selecting ROIs for users. IAMSAM offers users with two modes for running the SAM algorithm: everything-mode and prompt-mode.

• Everything-mode : An automatic mode that generates masks for the entire image, providing a comprehensive analysis of the spatial gene expression patterns.

• Prompt-mode : An interactive mode that allows users to guide the segmentation process by providing box prompts.

After selecting ROIs, IAMSAM automatically performs downstream analysis including identification of differentially expressed genes, enrichment analysis, and cell type prediction within the selected regions.