01 Open-source work
Python · GPU · ★ 29
A PyTorch reimplementation of RCTD for deconvolving spatial transcriptomics spots (Visium, Xenium, MERFISH, Slide-seq) into cell-type proportions. Drop-in replacement for the R spacexr package, 4–40× faster on GPU with 99.7% concordance on benchmarks. Works on CPU when a GPU isn’t available.
Deconvolution
PyTorch
Visium / Xenium / MERFISH
Curated list · ★ 139
A curated, opinionated index of the spatial transcriptomics tooling landscape — segmentation, cell-type annotation, deconvolution, differential expression, benchmarks, datasets, foundation models. Started as my own working notes; now community-maintained with regular updates as the field moves.
Survey
Segmentation
Benchmarks
Foundation models
Notebooks
Reproducible analysis notebooks accompanying my Science Immunology paper on tumor-infiltrating CD8 T cells in non-small-cell lung cancer. End-to-end from count matrices to the figures in the paper.
scRNA-seq
scTCR-seq
Reproducible