EXPlainable VAE for ORdinally perturbed Transcriptomics data (EXPORT) is an interpretable VAE model with a biological pathway informed architecture, to analyze ordinally perturbed transcriptomics data. Specifically, the low-dimensional latent representations in EXPORT are ordinally-guided by training an auxiliary deep ordinal regressor network and explicitly modeling the ordinality in the training loss function with an additional ordinal-based cumulative link loss term.
If you use this code, please cite our preprint paper:
Niyakan, S., Luo, X., Yoon, B. & Qian, X. (2024). Biologically Interpretable VAE with Supervision for Transcriptomics Data Under Ordinal Perturbations, bioRxiv. https://www.biorxiv.org/content/early/2024/03/29/2024.03.28.587231
In order to analyze your multi-sample spatial transcriptional (ST) data with MUSTANG, 4 main steps should be performed:
- Spots Spatial Graph: The adjacency matrix of spots spatial graph should be extracted based on the layout.
- Spots Transcriptional Graph: The adjacency matrix of spots transcriptional graph in which spots that are transcriptionally similar to eachother are connected with an edge should be extracted.
- Spots Similarity Graph: The adjacency matrix of spots similarity graph should be constructed based on adjacency matrices of spots spatial and transcriptional graphs.
- Bayesian Deconvolution Analysis: The Poisson discrete deconvolution model should be applied to extract the deconvolution parameters.