Installation

pip install step-kit

Then install dgl for your specific environment. For example, to install dgl 1.1.3 with cuda 11.7, which is the version used in the tutorials, you can run the following command:

pip install dgl==1.1.3 -f https://data.dgl.ai/wheels/cu117/repo.html

dgl official installation guide can be found here.

Usage

• EntryPoints
• Results Storage

Tutorials

This page contains tutorials on how to use the STEP.

• Spatial domain identification on 10x Visium DLFPC data
• Performing integrative analysis on scRNA-seq data and 10x Visium data of Human Lymph Node with STEP
• Reveal multiple level of biological heterogeneities in Mouse Hypothalamus by MERFISH-seq
• Integrate Human Immune Cell datasets with STEP and generate batch-corrected embeddings and gene expressions.
• Visium HD Human Colorectal Cancer (16 um) cell type clustering & spatial domain identification
• Visium HD Mouse Small Intestine (8 um and 16 um) cell type clustering & spatial domain identification

API Reference

This page contains auto-generated API reference documentation.

• step
  • step.functionality
    • step.functionality.comps
      • step.functionality.comps.model_ops
      • step.functionality.comps.trainer
    • step.functionality.base
    • step.functionality.cross_funcmodel
    • step.functionality.sc_funcmodel
    • step.functionality.st_funcmodel
  • step.manager
    • step.manager.save
  • step.models
    • step.models.clust
    • step.models.deconv
    • step.models.distributions
    • step.models.extension
    • step.models.knn_map
    • step.models.nnls
    • step.models.transcriptformer
  • step.modules
    • step.modules.decoder
    • step.modules.gnn
    • step.modules.transformer
  • step.utils
    • step.utils.synthetics
      • step.utils.synthetics.pseudo_st
    • step.utils.dataset
    • step.utils.gbolt
    • step.utils.metrics
    • step.utils.misc
    • step.utils.plotting
  • step.integration
  • step.scmodel
  • step.stmodel