Ecosystem interop

LeafletFA stores all results in standard AnnData slots, so the output integrates directly with the broader single-cell Python ecosystem.

scanpy

Cell activities in adata.obsm["X_PHI"] work as a drop-in replacement for any PCA/scVI embedding:

import scanpy as sc

sc.pp.neighbors(adata, use_rep="X_PHI")
sc.tl.umap(adata)
sc.tl.leiden(adata)
sc.pl.umap(adata, color=["leiden", "tissue", "age"])

scvi-tools

SpliceVI — the sister model — is built on scvi-tools and shares the same SplicingDataset input format. If you want a joint splicing + gene expression latent space, use SpliceVI; the adata.obsm["X_PHI"] slot name is compatible.

muon

For multimodal objects combining splicing with chromatin or gene expression modalities, muon can hold the SplicingDataset as one modality:

import muon as mu

mdata = mu.MuData({"splicing": adata_splicing, "rna": adata_rna})

pyroe / tximeta

If starting from alevin-fry or salmon quantification, pyroe can produce per-cell count matrices that ATSEmapper can consume after junction extraction with regtools.