cyclevi 0.1.2

CycleVI: Isolating cell cycle variation with an interpretable deep generative model

CycleVI

CycleVI is a deep generative model that isolates cell cycle variation in single-cell RNA-seq data. It learns a disentangled latent space where two dimensions capture circular cell cycle position (z_cycle) and the remaining dimensions capture everything else (z_other) — so downstream analyses are not confounded by cell cycle.

Preprint: CycleVI: Isolating cell cycle variation with an interpretable deep generative model

---

Installation

pip install cyclevi

To run without installing (from the repository root):

python -m cyclevi

Dependencies: PyTorch, scvi-tools, anndata, scanpy, scikit-learn, click, numpy, pandas.

---

Quickstart

One command runs the full pipeline — load data, compute phase initialization, train the model, save outputs:

cyclevi run --input data.h5ad --output results/

Outputs written to results/:

File	Contents
model/	Saved model (reload with CycleVI.load)
latent_cycle.csv	2D circular cell cycle coordinates (z_cycle_x, z_cycle_y)
latent_other.csv	Non-cycling latent dimensions (z_1 … z_N)
cycle_angles.csv	Inferred cell cycle angle per cell (radians, −π to π)

---

Step-by-step workflow

For more control, or to reuse the prepared file across multiple runs:

# Step 1: compute phase initialization, save a prepared file
cyclevi prepare --input data.h5ad --output prepared.h5ad

# Step 2: train (layer and column names are read from the file automatically)
cyclevi train --input prepared.h5ad --output results/

The prepared file stores all settings internally, so cyclevi train needs no extra flags.

---

Supported input formats

All commands that accept --input support the following formats:

Format	Description
.h5ad	AnnData HDF5 file
.h5	10x Genomics HDF5 file (Cell Ranger output)
.loom	Loom file (e.g. from velocyto)
.mtx	MatrixMarket sparse matrix (cells as rows by default)
directory	10x Genomics MTX folder (matrix.mtx, barcodes.tsv, features.tsv)
.csv / .tsv	Delimited text (cells as rows, genes as columns by default)

For .mtx and .csv/.tsv files where genes are rows, add --transpose.

---

Commands

cyclevi run

End-to-end pipeline in one step.

cyclevi run --input data.h5ad --output results/

Accepts all options from prepare and train combined (see below).

---

cyclevi prepare

Compute phase initialization and save a prepared .h5ad file.

CycleVI infers cell cycle position itself — this step provides the initial phase guesses the model needs to start training. It scores cells using known S and G2/M marker genes, derives a continuous angle from the scores, and quantile-transforms it so cells are spread uniformly around the circle.

cyclevi prepare --input data.h5ad --output prepared.h5ad

Options:

Option	Default	Description
--input	—	Input file or directory
--output	—	Path to save the prepared .h5ad file
--gene-id-type	auto	Gene identifier type: auto detects from var_names, or set ensembl / symbol explicitly
--var-names	gene_symbols	For 10x MTX directories: use gene_symbols or gene_ids as var_names
--transpose	off	Transpose matrix after loading (for MTX/CSV with genes as rows)
--counts-layer	counts	Layer name for storing a copy of raw counts
--phase-key	phase	adata.obs key for discrete phase labels (G1 / S / G2M)
--angle-key	cycle_angle	adata.obs key for the raw arctan2 angle
--uniform-angle-key	cycle_angle_uniform	adata.obs key for the quantile-transformed angle (used for training)

Columns added to adata.obs:

Column	Description
phase	Initial phase label: G1, S, or G2M
S_score	Continuous S-phase score
G2M_score	Continuous G2/M-phase score
cycle_angle	Raw angle: arctan2(G2M_score, S_score) in [0, 2π]
cycle_angle_uniform	Quantile-transformed angle, uniformly distributed in [0, 2π]

The prepared file stores these column names in adata.uns["cyclevi"] so that cyclevi train picks them up automatically.

---

cyclevi train

Train a CycleVI model from a prepared .h5ad file.

cyclevi train --input prepared.h5ad --output results/

If the file was created by cyclevi prepare, no additional flags are needed — the layer and column names are read from the file automatically.

Options:

Option	Default	Description
--input	—	Prepared .h5ad file
--output	—	Output directory
--batch-key	None	adata.obs column for experimental batch
--labels-key	None	adata.obs column for cell type labels
--cycle-label-key	auto	adata.obs column for phase labels — read from file if prepared with cyclevi prepare
--cycle-angle-key	auto	adata.obs column for cycle angle — read from file if prepared with cyclevi prepare
--layer	auto	AnnData layer with raw counts — read from file if prepared with cyclevi prepare
--n-latent	10	Total latent dimensions (first 2 are always z_cycle)
--n-hidden	128	Hidden units per encoder/decoder layer
--n-layers	1	Number of encoder/decoder layers
--n-epochs	400	Training epochs
--batch-size	128	Mini-batch size
--lr	1e-3	Learning rate

Outputs saved to --output:

File	Contents
model/	Saved model (reload with CycleVI.load)
latent_cycle.csv	2D circular cell cycle coordinates (z_cycle_x, z_cycle_y)
latent_other.csv	Non-cycling latent dimensions (z_1 … z_N)
cycle_angles.csv	Inferred cell cycle angle per cell (radians, −π to π)

---

cyclevi extract

Extract latent representations from a previously saved model without retraining.

cyclevi extract \
  --input data.h5ad \
  --model results/model \
  --output results/

Writes the same three CSV files as train.

Option	Description
--input	Input .h5ad file
--model	Directory of a saved CycleVI model
--output	Directory to write output CSVs

---

Python API

import anndata as ad
from cyclevi import CycleVI
from cyclevi.prepare import load_adata, compute_phase_initialization

# Load and prepare data
adata = load_adata("data.h5ad")
compute_phase_initialization(adata)

# Train
CycleVI.setup_anndata(
    adata,
    layer="counts",
    cycle_initiation_label_key="phase",
    cycle_initiation_angle_key="cycle_angle_uniform",
)
model = CycleVI(adata)
model.train(max_epochs=400)

# Extract latent representations
import numpy as np
z = model.get_latent_representation(adata)
z_cycle   = z[:, :2]          # circular cell cycle coordinates
z_other   = z[:, 2:]          # non-cycling dimensions
angles    = np.arctan2(z[:, 1], z[:, 0])  # cell cycle angle

For a full walkthrough, see Tutorial.ipynb or Tutorial_colab.ipynb (Google Colab ready).

---

Feedback

For questions and comments, contact Gustavo S. Jeuken or open an issue on GitHub.

License

BSD 3-Clause License

Citation

If you use CycleVI in a publication, please cite:

CycleVI: Isolating cell cycle variation with an interpretable deep generative model

Pia Mozdzanowski, Marcel Tarbier, Gustavo S. Jeuken

bioRxiv 2025.11.04.686009; doi: https://doi.org/10.1101/2025.11.04.686009