A crude (for now) implementation of various Gaussian process models for spatial transcriptomics data. Currently planned to contain implementations of:
FaST-LMM [Lippert et al.]:
- LMMs/GPs with full-rank kernel matrix
- LMMs/GPs with low-rank kernel matrix factorization
SpatialDE [Svensson et al.]:
- Variance parameter estimation in SpatialDE model
- Statistical significance of spatial vs. non-spatial covariance
- Expression histology
Spatial Variance Component Analysis (SVCA) [Arnol et al.]:
- Variance parameter estimation in SVCE model
I'm also hoping to implement the method of Greengard et al. to see if it leads to further speed-up.