SCIMAP Pro¶
Spatial single-cell analysis for AnnData and SpatialData.
SCIMAP Pro (scimappro) is the professional edition of
SCIMAP. It keeps the same analytical vocabulary —
phenotyping, neighbourhood statistics, spatial distance, co-occurrence,
proximity scores — and rebuilds it around three things the community edition
does not offer: native scverse SpatialData
support, out-of-core streaming for .h5ad files that do not fit in memory, and
vectorised, parallel implementations of the heavy spatial kernels.
import scimappro as sp
adata = sp.pp.mcmicro_to_scimap("quantification/unmicst-exemplar.csv")
adata = sp.pp.rescale(adata, gate="manual_gates.csv")
adata = sp.tl.phenotype(adata, phenotype="phenotype_workflow.csv")
adata = sp.tl.spatialCooccurrence(adata, phenotype="phenotype")
sp.pl.spatialCooccurrence(adata)
One data argument¶
Every public function takes its cell table as the first argument, named data,
and accepts an AnnData, an .h5ad path, a SpatialData, or a .zarr store.
There is no separate SpatialData API to learn.
Read more
Streams what will not fit¶
streamData=True runs the analysis through
CAP-AnnData, reading only the
obs columns, markers, and matrix slices a function actually needs, and writing
back only the sections that changed.
Read more
Built for large slides¶
BallTree neighbourhood queries, Polars group-bys, Numba permutation kernels,
and maxWorkers-controlled parallelism over images, in place of the row-wise
Pandas loops of the original implementation.
Where to start¶
| If you want to… | Go to |
|---|---|
| Install the package | Install |
| See a whole analysis end to end | Quickstart |
| Move an existing scimap script across | Migrating from scimap |
| Look up a function's parameters | API index |
| Work through a real dataset | Tutorials |
Relationship to SCIMAP¶
SCIMAP Pro is a separate package with a deliberately incompatible API —
adata became data, snake_case became camelCase, and several functions were
renamed to say what they measure. Nothing imports scimap, and the two can be
installed side by side. The migration guide maps every renamed
function and parameter.
The community edition remains available and documented at scimap.xyz.
Citation¶
If SCIMAP Pro contributes to work you publish, please cite the SCIMAP paper:
Nirmal et al., (2024). SCIMAP: A Python Toolkit for Integrated Spatial Analysis of Multiplexed Imaging Data. Journal of Open Source Software, 9(97), 6604. https://doi.org/10.21105/joss.06604
SCIMAP development was led by Ajit Johnson Nirmal, Harvard Medical School. See the other tools from the Nirmal Lab.
Funding¶
This work was supported by NIH grant K99-CA256497.
License¶
Scimappro Academic License 1.0 — free for academic research, teaching, and other noncommercial use. Redistribution of modified copies is not permitted, and commercial use requires a separate license. See License.