multiDGD - processed data and models

multiDGD is a new generative model for multi-omics data. This repository contains processed single-cell multiome datasets and trained models used in multiDGD paper (multiDGD package) (analysis repo)Dataset IDshuman_bonemarrow - bone marrow dataset from NeurIPS OpenProblems competitionhuman_brain - Human brain dataset from 10X genomicsmouse_gastrulation - Mouse gastrulation dataset from Argelaguet et al. 2022File types{dataset_id}.h5ad - AnnData object of processed dataset{dataset_id}.h5mu - MuData object of processed datasetdgd_{dataset_id}.pt - trained multiDGD modeldgd_{dataset_id}_hyperparameters.json - model hyperparameters used during training