API

Import SC2Spa:

import SC2Spa

Spatial Inference

tl.FineMapping(adata_ref, adata_query[, ...])	Finely map single cells to spatial locations and Reconstruct ST data at single cell resolution
tl.Reconstruct_scST(adata_ref, adata_query)	Reconstruct ST data at single cell resolution
tl.NRD_CT_preprocess(adata_ref, adata_query)	Apply KNN algorithm to obtain the single cell neighbors of ST beads
tl.NRD_weight(neighbors, dis, adata_ref, ...)	Calculate the weights of nearby single cells for ST beads based on the fine mapping result
tl.NRD_CT(neighbors, dis, adata_ref, adata_query)	Normalized Reciprocal Distance
tl.NRD_impute(neighbors, dis, adata_ref, ...)	Reconstruct the gene expression profile of ST beads based on the NRD_weight output
tl.Train_transfer(adata, root, model_root[, ...])	Finetune location prediction model (FCNN) on a specific cell type A model will be trained and saved to root + 'SI_' + CT + '.h5' The predicted coordinates of single cells will be saved in adata_query.obsm['spatial_mapping'] The predicted coordinates of beads will be saved in adata_ref.obs['spatial_mapping'] Fine mapping information will be saved in adata_ref.obs['FM'] and adata_query.obs['FM'].
tl.SaveValidation(history, CV, name)	Save the FCNN or LR training history to 'log/training_log_' + name + '.pickle',
tl.CheckAccuracy(name[, item_name])	Check the accuracies and mean accuracy of cross-validation
tl.CrossValidation(X, Y, train_indices, ...)	Perform Cross-validation using fully connected neural network
tl.WassersteinD(adata_ref, adata_query, ...)	Calculate Wassertein distances of genes between two datasets
tl.pp_Mapping(adata_ref, adata_query, JGs[, ...])	Extract gene expression matrices sharing the same genes and the reference coordinates
tl.BatchPredict(Model, X[, BatchSize])	Predict batch by batch

Plotting

pl.DrawGenes2(adata, gene[, coords_name, ...])	Show the expression of a gene in cells or ST beads
pl.DrawCT1(adata, CT[, ax, coords_name, s, ...])	Show the spatial locations of a type of cells or beads
pl.DrawCT2(adata, CT[, coords_name, title, ...])	Customized Function
pl.DrawCT3(adata, CT_list[, coords_name, s, ...])	Display original locations or predicted locations of beads/cells of selected types.
pl.DrawSVG(adata, GeneList, target_field[, ...])	Show the expression of location predictive genes (or spatially variable genes)
pl.Superimpose(adata[, coords_name, G1, G2, ...])	Show the spatial gene expressions of two genes and their superimposed images
pl.draw_cb([cmap, figsize, save, size])	Draw an independent color bar figure

Benchmarking

bm.BVMI(adata1, adata2, GL[, coords_name, ...])	Calculate bivariate Moran's I of the genes of two ST data.
bm.Vis_Euclidean(coord_ref, coord_pred, ...)	Show the euclidean distance of the original and predicted locations of beads.

Preprocessing

pp.MinMaxNorm(Y, Y_ref)	Min-Max normalize along the second axis
pp.ReMMNorm(Y_ref, Y_pred)	Reverse Min-Max normalize along the second axis
pp.PolarTrans(Y)	Transform cartesian coordinates to polar coordinates
pp.RePolarTrans(RTheta)	Transform polar coordinates to cartesian coordinates

Spatially Variable Gene Analysis

sva.PrioritizeLPG(adata, Model[, sparse, ...])	Prioritize genes' contribution to location prediction
sva.SelectGenes(Weights[, percent])	Trace back the weight matrices to evaluate the importance of genes in location prediction
sva.SelectFeatures(array_in, percent)	Select top percent quantile nodes

Mutual Exclusivity Analysis

me.BME_stat(A1, A2[, p_coef])	Calculate Balanced Mutual exclusivity score between Gene 1 and Gene 2
me.calc_BME(A[, n_process, n_resamples])	Calculate Balanced Mutual Exclusivity
me.calc_BME_sub(ij, A, genes, n_resamples)	Calculate Balanced Mutual exclusivity (subprocess)
me.calc_DEEI(A[, cutoff, p_coef, n_process])	Calculate Balanced Mutual Exclusivity and Directed exclusively express index
me.DEEI_sub(ij, Count_excl, Prob_excl, A_S, ...)	Calculate Balanced Mutual exclusivity and Directed exclusively express index (subprocess)
me.Count_Prob_EEI(A)	Calculate the counts and probabilities of exclusive events