API

Import ncem as:

import ncem

Tools

NCEM tools containing linear models, variance decomposition and ablation study.

tl.linear_ncem(adata, key_type, key_graph[, …])	Fit a linear NCEM based on an adata instance and save fits in instance.
tl.linear_ncem_deconvoluted(adata, …[, …])	Fit a linear NCEM based on deconvoluted data in an adata instance and save fits in instance.
tl.differential_ncem(adata, …[, formula, …])	Fit a differential NCEM based on an adata instance and save fits in instance.
tl.differential_ncem_deconvoluted(adata, …)	Fit a differential NCEM based on deconvoluted data in an adata instance and save fits in instance.
tl.spline_linear_ncem(adata, df, …[, …])	Fit a linear NCEM based on an adata instance and save fits in instance.
tl.spline_linear_ncem_deconvoluted(adata, …)	Fit a linear NCEM based on deconvoluted data in an adata instance and save fits in instance.
tl.spline_differential_ncem(adata, df, …)	Fit a differential NCEM based on an adata instance and save fits in instance.
tl.spline_differential_ncem_deconvoluted(…)	Fit a differential NCEM based on deconvoluted data in an adata instance and save fits in instance.

Plotting

NCEM tools containing plotting functions.

pl.cluster_freq(adata, cluster_key[, title, …])	Plot cluster frequencies.
pl.noise_structure(adata, cluster_key[, …])	Plot cluster frequencies.
pl.circular(adata, alpha, scale_edge[, …])	Plot cluster frequencies.
pl.circular_rotated_labels(adata, alpha, …)	Plot cluster frequencies.
pl.ablation(adata[, figsize])	Plot of ablation study results

Model classes: models

Model classes from ncem for advanced use.

Classes that wrap tensorflow models.

models.ModelCVAE(input_shapes, latent_dim, …)	Model class for conditional variational autoencoder.
models.ModelCVAEncem(input_shapes, …)	Model class for NCEM conditional variational autoencoder with graph layer IND (MAX) or GCN.
models.ModelED(input_shapes, latent_dim, …)	Model class for non-spatial encoder-decoder.
models.ModelEDncem(input_shapes, latent_dim, …)	Model class for NCEM encoder-decoder with graph layer IND (MAX) or GCN.
models.ModelInteractions(input_shapes, …)	Model class for interaction model, baseline and spatial model.
models.ModelLinear(input_shapes, l2_coef, …)	Model class for linear model, baseline and spatial model.