Weghted Ensemble method of Spatial Transcriptomics (WEST)

1. Introduction

WEST is a method that utilizes ensemble techniques to improve the performance and stability of spatial transcriptomics data analysis. It represents a significant advance in clustering spatial transcriptomics data, offering improved accuracy and flexibility compared to existing methods, making it a valuable tool for spatial transcriptomics data analysis.

2. Environment

Python=3.8

Required packages:

The detailed requirements can be found here.

3.Tutorial

The tutorial provides a detailed tutorial on how to implement WEST on an actual sample, including the description for every parameter. The input of WEST is the embeddings generated by other deep learning-based methods. The tutorials on implementing the six benchmarking methods: Leiden, SpaGCN, SpaceFlow, SEDR, STAGATE, and DeepST are also provided. Support resouce provides some required code for the benchmarking methods.

4. Data used in the paper

4.1 Simulation data

The code for simulation data is provided. Square is the code for generating simulation data with the squared spatial pattern, and circle is the code for generating simulation data with the circled spatial pattern.

4.2 Read data

The DLPFC data used in the paper can be found here.
The mouse embryo data used in the paper can be found here.
The 10X Genomics Visium fluorescence image data of the mouse brain used in the paper can be found here. The 10X Genomics Visium H&E image data of the mouse brain used in the paper can be found here. The human pancreatic cancer data used in the paper can be found here.