Package index • tAge

Pipelines

End-to-end entry points. tAge_preprocessing() returns the list of ExpressionSets that predict_tAge() consumes.

tAge_preprocessing(): Complete preprocessing pipeline for tAge analysis
predict_tAge(): Predict transcriptomic age with several clocks
predict_tAge_one(): Predict transcriptomic age with one pre-trained model
tAge_by_group(): Run tAge pipeline separately per group factor (e.g., "tissue")

Clock catalogue

Browse the registry of pre-trained models and fetch them from Zenodo.

list_clocks(): List available transcriptomic clock models
list_module_clocks(): List available module clock models
download_clocks(): Download clock models from Zenodo
tage_species(): Species supported by the clocks

Preprocessing steps

The individual stages run by tAge_preprocessing(), for when you need a non-standard order or want to inspect an intermediate.

filter_genes(): Filter genes based on expression thresholds
map_genes(): Map genes in an ExpressionSet using local CSV mapping tables
RLE_normalization(): Perform RLE (Relative Log Expression) normalization
log_transform(): Apply log10 transformation to expression data
scale_eset(): Scale expression data to have zero mean and unit variance
YuGene(): Perform YuGene normalization
control_subtraction(): Subtract reference group median from expression data (relative expression)

Single-cell and pseudobulk

Pool single cells into pseudobulk samples, drop outliers, and move data in from Seurat or h5ad.

aggregate_pseudobulk(): Aggregate single-cell data into pseudobulk samples based on read coverage
aggregate_on_obs_columns(): Aggregate single-cell data into pseudobulk samples within obs column groups
pseudobulk_summary(): Get summary statistics for pseudobulk samples
flag_outliers(): Flag outlier samples
load_h5ad_simple(): Load h5ad file using reticulate and anndata
load_h5ad_to_seurat(): Load AnnData h5ad file and convert to Seurat object
subset_seurat_by_metadata(): Subset Seurat object by metadata criteria

Statistics

The group comparisons of the TACO / tClock application: marginal-mean contrasts for elastic net clocks, REML meta-regression for Bayesian ridge clocks, module-clock effect sizes.

tage_compare_groups(): Compare tAge between experimental groups
tage_regress_continuous(): Regress tAge on a continuous predictor
tage_module_stats(): Module-clock effect sizes and p-values
tage_adjust_covariates(): Covariate-adjusted tAge values for plotting
tage_significance_stars(): Significance stars for tAge statistics

Plotting

Figures built directly on the statistics, plus QC plots.

tage_clock_forest(): Forest plot of clock effects
tage_module_heatmap(): Heatmap of module-clock effects
tage_boxplot(): Box plot of tAge predictions with pairwise significance annotation
plot_eset_density(): Plot density curves for ExpressionSet data
load_module_functions(): Module colour to biological function map
tage_fig_size(): Size a tAge figure was designed for
tage_save_plot(): Save a tAge figure at the size it was designed for
theme_tage(): The tAge ggplot2 theme
tage_colors(): The tAge colour tokens
tage_series_colors(): Colour per group level
scale_fill_tage_diverging(): Diverging fill scale for signed effects

Data and helpers

Build an ExpressionSet and reach the example data shipped with the package.

make_ExpressionSet(): Create ExpressionSet object from expression data and phenotype data
load_example_expression_data(): Load example expression data
load_example_metadata(): Load example metadata
load_gene_list(): Load gene list
get_package_data(): Get package data file paths
get_metadata_dir(): Get path to the package metadata directory