RustScenic

Faster, memory-efficient regulatory-network analysis for single-cell and multiome data.

Rust kernels for GRN inference, regulon activity, motif enrichment, topic modelling, enhancer links and eRegulons. Python API. CPU-first. One install.

Documentation | Benchmarks | Validation | Citation | Zenodo DOI

Highlights

• 11x to 52x faster than SCENIC+ in tested real-data core E2E rows
• 6.34 GB peak RSS on a 100k-cell four-stage scale check; legacy pySCENIC reports exceed 40 GB on similar workloads
• Current release: v0.4.7
• pip install rustscenic, with Python 3.10 to 3.13 release wheels
• Huang Lab collaborator run recovered 16/17 expected brain TFs on 10x human brain GEM-X data
• Rust implementations for the matrix-heavy regulatory-network stages
• Core path runs without Java, dask, CUDA or Snakemake
• Benchmark artefacts include commands, hardware, runtime, memory and output checks

Installation

pip install rustscenic

Benchmark Evidence

Core E2E comparison on the same matrix-level regulatory path: TF-to-gene, region-to-gene, eRegulons, gene AUCell and region AUCell.

This is a practical output-path benchmark against SCENIC+. It is not a claim that every internal stage uses the same estimator: RustScenic enhancer linking uses correlation over the fixed search space, while SCENIC+ uses GBM plus Pearson scoring for region-to-gene links.

Machine: Apple M5 laptop, 16 GB RAM, macOS arm64, 4 CPU threads. RustScenic rows used Python 3.13.9; SCENIC+ reference rows used Python 3.11.8 for its dependency stack. Rows can be sampled subsets; the shape column is the actual benchmark input.

Dataset	Shape	RustScenic	SCENIC+	Speedup	Peak RSS (RustScenic / SCENIC+)
PBMC3k dense	2,000 cells, 4,000 genes, 8,000 peaks, 30 TFs	4.98 s	258.9 s	52x	1.21 / 1.26 GB
PBMC10k dense	2,000 sampled cells, 4,000 genes, 8,000 peaks, 30 TFs	21.5 s	241.5 s	11x	2.37 / 2.63 GB
Mouse brain E18	1,500 cells, 3,000 genes, 6,000 peaks, 25 TFs	2.82 s	90.4 s	32x	1.65 / 2.10 GB
Human brain GEM-X	2,000 cells, 4,000 genes, 8,000 peaks, 30 TFs	7.41 s	146.0 s	19.7x	2.18 / 2.19 GB

Including data preparation, the human brain GEM-X row is 11.89 s for RustScenic versus 150.36 s for SCENIC+.

Full commands, hardware, validation metrics and output signatures are in site_docs/benchmarks.md.

Stage Coverage

Stage	RustScenic API	SCENIC ecosystem stage covered
TF-to-gene GRN	rustscenic.grn.infer	GRNBoost2-style regulatory-network inference
AUCell	rustscenic.aucell.score	Per-cell regulon activity scoring
cisTarget	rustscenic.cistarget.enrich	Motif enrichment and support filtering
Topics	rustscenic.topics.fit, fit_gibbs	scATAC topic modelling
ATAC preprocessing	rustscenic.preproc	Fragment matrix building and QC
Enhancer links	rustscenic.enhancer.link_peaks_to_genes	Peak-to-gene linking
eRegulons	rustscenic.eregulon.build_eregulons	Enhancer-linked regulon assembly
Orchestration	rustscenic.pipeline.run	Staged workflow across RNA and multiome inputs

Quick Start

import anndata as ad
import rustscenic.aucell
import rustscenic.data
import rustscenic.grn

adata = ad.read_h5ad("rna.h5ad")
tfs = rustscenic.data.tfs("hs")

grn = rustscenic.grn.infer(adata, tf_names=tfs, n_estimators=5000, seed=777)

regulons = [
    (f"{tf}_regulon", grn[grn["TF"] == tf].nlargest(50, "importance")["target"].tolist())
    for tf in grn["TF"].unique()
]
auc = rustscenic.aucell.score(adata, regulons, top_frac=0.05)

Command line:

rustscenic pipeline --rna data.h5ad --tfs tfs.txt --output out/
rustscenic grn --expression rna.h5ad --tfs tfs.txt --output grn.parquet
rustscenic aucell --expression rna.h5ad --regulons grn.parquet --output aucell.parquet
rustscenic topics --expression atac.h5ad --output topics.parquet --n-topics 30
rustscenic cistarget --rankings rankings.feather --regulons regulons.tsv --output motifs.parquet

See examples/pbmc3k_end_to_end.py for a small real-data RNA example.

Validation

Validation axis	Result
cisTarget kernel	Pearson 1.0000 against ctxcore.recovery.aucs; mean absolute difference about 2.4e-5.
AUCell parity	Ziegler 2021 airway atlas mean per-cell Pearson 0.984; 91.7% of cells above 0.95.
Human brain GEM-X benchmark	Region-to-gene Jaccard 1.000; region AUCell mean Pearson 0.823.
Collaborator lab artefact	10x human brain multiome full monolith run recovered 16/17 expected brain TFs.
Open parity targets	Gene AUCell Pearson 0.386 and eRegulon edge Jaccard 0.161 on the human brain GEM-X row.

Validation artefacts live under validation/. Public interpretation lives in site_docs/benchmarks.md and site_docs/validation.md.

Current Boundaries

• The headline benchmark is the core matrix-level E2E path, not every possible raw-fragment and motif-database workflow.
• GRN, gene AUCell and eRegulon edge agreement are not claimed to be bit-identical to SCENIC+; see Benchmarks for the parity metrics.
• Larger repeated real-data runs and second-machine measurements are the next benchmark tier.

Documentation

• Installation
• Quickstart
• API map
• Benchmarks
• Validation
• Scope

Citation

If you use RustScenic in a paper, report, benchmark, derivative package or lab workflow, cite the exact release used. GitHub citation metadata is in CITATION.cff. Zenodo concept DOI: 10.5281/zenodo.20246040.

RustScenic is created and maintained by Ekin Kahraman. See AUTHORS.md for attribution.