Semi-Supervised Deconvolution of Bulk RNA-Seq Data with Hyperparameter Optimization

---

Documentation for package ‘dicepro’ version 1.0.2

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

dicepro-package	dicepro: Semi-Supervised Deconvolution of Bulk RNA-Seq Data with Hyperparameter Optimization
best_hyperParams	Select optimal hyper-parameters using a Pareto frontier
BlueCode	BlueCode Reference Signature Matrix
CellMixtures	CellMixtures Bulk RNA-seq Dataset
contains_nan_or_inf	Check if a value contains NaN or Inf
create_gamma_lambda_plot	Visualize the gamma–lambda hyper-parameter search space
dicepro	Semi-supervised bulk RNA-seq deconvolution with hyper-parameter optimization
full_metrics	Full agreement metrics via mixed-effects modeling
generateProp	Generate Cell-Type Proportion Matrix
generate_ref_matrix	Generate a Reference Signature Matrix
heatmap_abundances	Heatmap of cell-type abundances
makeTable1Tool	Build a performance metrics table for composition matrices
metric_plot	Performance metric line plot across iterations
nmf_lbfgsb	NMF with L-BFGS-B optimization
nmf_lbfgsb_hyperOpt	NMF L-BFGS-B wrapper for hyper-parameter optimization
nrmse	Normalized Root Mean Square Error (NRMSE)
objective_opt	Objective function for hyper-parameter optimization
plot.dicepro	Plot cell abundance heatmap and error plot
plot_hyperopt	Plot hyperparameter optimization report
plot_hyperopt.dicepro	Plot hyperparameter optimization report
research_hyperOpt	Hyper-parameter optimization loop for dicepro
row_norm_pos	Row-normalize a matrix with non-negative clamping
running_method	Run cell-type deconvolution
run_CSx	Run CIBERSORTx Deconvolution Method
run_experiment	Run a dicepro hyperparameter optimization experiment
samplewise_metrics	Sample-wise Pearson correlation and RMSE
simulation	Simulate Bulk RNA-seq Data with Biological and Technical Noise
simulation_bluecode	Simulate Bulk RNA-seq Data Using the BlueCode Reference Matrix