Package index • OmicSelector

Single-sample scoring & deployment

Freeze rostered within-sample scorers and score one incoming specimen without a test batch; deployability only, not a superiority claim.

deploy_singlesample(): Build a frozen single-sample deployment scorer
score_specimen(): Score new specimens with a frozen single-sample deployment object
is_singlesample_deployable(): Check whether a deployment scorer is single-sample deployable
ws_balance_ilr(): Isometric log-ratio balances on a frozen partition tree
ws_rclr_trimmed(): Robust trimmed centred log-ratio for compositional miRNA panels
singlesample_method_roster(): Amendment #4 method roster (frozen expansion)
singlesample_score_call(): Score a roster method with the canonical single-sample interface
singlesample_assert_row_equivariant(): Assert that a scorer is row-equivariant (single-sample deployable)

Core Pipeline

Main classes for building zero-leakage ML pipelines

OmicPipeline: OmicPipeline: Zero-Leakage Feature Selection Pipeline
omic_pipeline(): Quick pipeline creation from data
BenchmarkService: BenchmarkService: Nested Cross-Validation with Zero Leakage
omic_benchmark(): Create benchmark service from OmicPipeline

Signature Selection

Multi-objective biomarker signature selection

select_best_signature(): Select Best Biomarker Signature from Nested CV Results
get_consensus_features(): Get Consensus Features from Best Signature
get_selected_features_per_fold(): Get Selected Features Per Fold

Stability Analysis

Feature selection stability metrics

compute_nogueira_stability(): Compute Nogueira Stability Index
compute_stability_from_resample(): Compute Stability from ResampleResult
extract_features_from_resample(): Extract Features from All Folds in ResampleResult
extract_selected_features(): Extract Selected Features from Trained GraphLearner

Batch Correction

FrozenComBat for proper batch effect handling

FrozenComBat: FrozenComBat R6 Class
frozen_combat_correct(): Convenience function for frozen ComBat correction
create_frozen_combat_pipeop(): Create a Frozen ComBat PipeOp for mlr3pipelines

Cross-Platform Transfer

Cross-platform model deployment and domain adaptation

cross_platform_transfer(): Convenience Wrapper for Cross-Platform Transfer

Within-Sample CoDA Methods

Paper 1 v2.2 image encodings, neural learners, and perturbation benchmarking

encode_simple_grid(): Simple Row-Major Grid Encoding
encode_corr_grid(): Encode with Correlation-Ordered Grid
encode_deepinsight(): Encode with DeepInsight Layout
encode_ratio_image(): Create Pairwise Log-Ratio Image
train_ratio_cnn(): Train Ratio Image CNN
train_ratio_cnn_multiseed(): Train Ratio CNN Across Three Default Seeds
clr_transform(): Centered Log-Ratio Transformation
train_clr_mlp(): Train a CLR + MLP Classifier
predict_clr_mlp(): Predict with a CLR + MLP Classifier
train_codacore(): Train a CoDaCoRe Classifier
predict_codacore(): Predict with a CoDaCoRe Classifier
ws_perturbation_benchmark(): Within-Sample Perturbation Benchmark

Calibration

Probability calibration and diagnostics

fit_platt_scaling(): Platt Scaling (Logistic Calibration)
fit_isotonic_calibration(): Isotonic Regression Calibration
fit_temperature_scaling(): Temperature Scaling
compute_ece(): Compute Expected Calibration Error (ECE)
decompose_brier(): Decompose Brier Score
calibration_summary(): Calibration Summary for Model Results
reliability_diagram_data(): Create Reliability Diagram Data

Interpretability

Model interpretation and feature importance

create_explainer(): Create Model Explainer
shap_values(): Compute SHAP-like Values
feature_importance(): Compute Permutation Feature Importance
partial_dependence(): Compute Partial Dependence
check_feature_correlations(): Check Feature Correlations

Phase 5: GOF Filters

Goodness-of-fit filters for sparse/zero-inflated omics data

FilterGOF_KS: Kolmogorov-Smirnov GOF Filter
FilterHurdle: Hurdle Filter for Zero-Inflated Data
FilterZeroProp: Zero-Proportion Filter
make_gof_filter(): Create GOF Filter
compare_gof_filters(): Compare GOF Filters on Task
register_gof_filters(): Register GOF Filters in mlr3

Phase 5: Bayesian Tuning

Bayesian hyperparameter optimization with mlr3mbo

make_autotuner_glmnet(): Create Bayesian-Optimized AutoTuner for glmnet
make_autotuner_xgboost(): Create Bayesian-Optimized AutoTuner for XGBoost
make_autotuner_ranger(): Create Bayesian-Optimized AutoTuner for Random Forest
make_autotuner_lightgbm(): Create Bayesian-Optimized AutoTuner for LightGBM
get_optimal_params(): Get Optimal Hyperparameters from AutoTuner
run_bayesian_benchmark(): Run Bayesian Optimization Benchmark

Phase 5: AutoXAI

DALEX-based interpretability with correlation warnings

xai_pipeline(): Run Complete XAI Pipeline
xai_explainer_mlr3(): Create DALEX Explainer from mlr3 Learner
xai_importance(): Compute Permutation Feature Importance
xai_correlations(): Compute Correlation Diagnostics for Features
xai_pdp(): Compute Partial Dependence Plots
xai_shap(): Compute SHAP Values for Observations
plot_xai_importance(): Plot XAI Feature Importance
print_xai_summary(): Print XAI Summary

Phase 5: Stability Ensemble

Bootstrap-based feature selection stability

create_stability_ensemble(): Create a Stability Ensemble with Presets
StabilityEnsemble: Stability-Based Ensemble Selection
SequentialSelector: Hybrid Sequential Feature Selection (HSFS)

Phase 5: Synthetic Data

Data augmentation and synthetic data generation

smote_augment(): SMOTE Augmentation for Omics Data
noise_augment(): Gaussian Noise Augmentation
validate_synthetic(): Validate Synthetic Data Quality
balance_classes(): Create Balanced Training Set
tabddpm_generate(): TabDDPM Synthetic Data Generator

Phase 5: Deep Learning

Deep learning models for omics (requires torch)

create_mlp_learner(): Create MLP Learner via mlr3torch
make_mlp_learner(): Make MLP Learner (Alias)
make_fttransformer_learner() make_tabtransformer_learner(): Create an FT-Transformer Learner
make_gnn_learner(): Create GNN Learner for Pathway-Aware Classification
build_correlation_adjacency(): Create Correlation-Based Adjacency for GNN
run_dl_benchmark(): Deep Learning Benchmark
check_dl_availability(): Check Deep Learning Availability

Multi-Omics

Multi-omics late integration

merge_omics_data(): Merge Multi-Omics Data for Analysis
get_modality_info(): Get Modality Information
validate_omics_input(): Validate Multi-Omics Input
stack_omics(): Create Multi-Omics Stacked Ensemble (Convenience Function)

Model Export & Reporting

Model deployment and TRIPOD reporting

generate_tripod_report(): Generate TRIPOD+AI Report
create_report_data(): Create Report Data Schema

Utilities

Helper functions for parallel processing and caching

setup_parallel(): Configure Parallelization for OmicSelector
with_parallel(): With Parallel Scope
get_parallel_status(): Get Current Parallelization Status
reset_parallel(): Reset Parallelization to Sequential
create_omic_cache(): Create Split-Aware Cache
cached_filter(): Cached Filter Computation
cache_stats(): Get Cache Statistics
clear_cache(): Clear OmicSelector Cache

Legacy Functions

Functions from OmicSelector 1.x (use with caution)

OmicSelector_correlation_plot(): OmicSelector_correlation_plot
OmicSelector_profileplot(): OmicSelector_profileplot
OmicSelector_propensity_score_matching(): OmicSelector_propensity_score_matching
OmicSelector_vulcano_plot(): OmicSelector_vulcano_plot

Internal Functions

Internal functions and helpers (exported for advanced use)

AutoXAI: Auto XAI: Automatic Explainability for Biomarker Models
BayesianTuner: Bayesian Hyperparameter Optimization for Omics
BenchmarkService: BenchmarkService: Nested Cross-Validation with Zero Leakage
CrossPlatformAdapter: CrossPlatformAdapter R6 Class
DeepLearners: Deep Learning Learners for Omics Data
FilterCoDA_CoDaCoRe: CoDaCoRe Feature Filter
FilterCoDA_LogcontrastLasso: Log-Contrast Lasso Filter
FilterCoDA_PLRVariance: Pairwise Log-Ratio Variance Filter
FilterCoDA_Selbal: Selbal-Style Forward Balance Filter
FilterCoDA_StabilityLogratio: Stability Log-Ratio Filter
FilterGOF: Goodness-of-Fit Filters for Sparse Omics Data
FilterGOF_KS: Kolmogorov-Smirnov GOF Filter
FilterHurdle: Hurdle Filter for Zero-Inflated Data
FilterZeroProp: Zero-Proportion Filter
FrozenComBat: FrozenComBat R6 Class
GoldStandard: Gold Standard Synthetic Dataset for Leakage Detection
OmicModalitySpec: OmicModalitySpec R6 Class
OmicPipeline: OmicPipeline: Zero-Leakage Feature Selection Pipeline
OmicSelector-data detectable_in_serum miRNAselector_tutorial_balanced_benchmark miRNAselector_tutorial_balanced_dataset miRNAselector_tutorial_balanced_mixed original_TCGA_data orginal_TCGA_data: Datasets included with OmicSelector
OmicSelector-package OmicSelector: OmicSelector: Zero-Leakage Biomarker Discovery Toolkit
OmicSelector_correlation_plot(): OmicSelector_correlation_plot
OmicSelector_profileplot(): OmicSelector_profileplot
OmicSelector_propensity_score_matching(): OmicSelector_propensity_score_matching
OmicSelector_vulcano_plot(): OmicSelector_vulcano_plot
OmicStackedEnsemble: OmicStackedEnsemble R6 Class
OmicWeightedEnsemble: Simple Weighted Averaging Ensemble
SequentialSelector: Hybrid Sequential Feature Selection (HSFS)
StabilityEnsemble: Stability-Based Ensemble Selection
SyntheticData: Synthetic Data Generation for Omics
TabularDL: Modern Tabular Learners for OmicSelector
apply_compositional_mahalanobis(): Apply compositional Mahalanobis distance to new samples
apply_frozen_combat_cv(): Apply Frozen ComBat Within Cross-Validation Folds
apply_frozen_quantile(): Apply frozen quantile calibration to new samples
apply_frozen_ruv(): Apply frozen RUV correction to new samples
apply_hemolysis_prefilter(): Apply a Hemolysis Pre-Filter
apply_hemolysis_rr(): Apply frozen hemolysis-RR correction to new samples
apply_isolation_forest_logratio(): Apply isolation forest to new samples and flag anomalies
apply_logistic_normal_eb(): Apply frozen logistic-normal EB shrinkage to new samples
apply_mirna_aliases(): Rename miRNA features in a matrix or named vector
apply_robust_pca_residual(): Apply MAD-scaled SVD residual filter to new samples
apply_sinkhorn_ot_scorer(): Apply a frozen Sinkhorn OT scorer.
autoencoder: Autoencoder Utilities (torch)
autoencoder_encode(): Encode Features with Autoencoder
autoencoder_fit(): Fit Autoencoder
autoencoder_load(): Load Autoencoder State
autoencoder_save(): Save Autoencoder State
balance_classes(): Create Balanced Training Set
bias-audit: Cohort-Provenance Bias Audit
biofluid_aware_scoring: Biofluid-aware within-sample scoring methods
build_correlation_adjacency(): Create Correlation-Based Adjacency for GNN
cache: Split-Aware Caching for OmicSelector
cache_stats(): Get Cache Statistics
cached_filter(): Cached Filter Computation
calibration: Calibration Metrics for OmicSelector
calibration_summary(): Calibration Summary for Model Results
check_batch_correction_leakage(): Check for Batch Correction Leakage
check_dl_availability(): Check Deep Learning Availability
check_feature_correlations(): Check Feature Correlations
check_no_premature_imputation(): Detect Global Pre-Loop Imputation (Leakage Check)
check_null_benchmark_draws(): Validate Random-Panel Null Benchmark
clear_cache(): Clear OmicSelector Cache
clr-mlp: CLR + MLP Models for Within-Sample Classification
clr_transform(): Centered Log-Ratio Transformation
coda-feature-selection: CoDA-Aware Feature Selection for Biomarker Panels
codaFS_codacore_wrapper(): CoDaCoRe-Derived Feature Selection
codaFS_logcontrast_lasso(): Log-Contrast Lasso on CLR Features
codaFS_plr_variance(): Pairwise Log-Ratio Variance Feature Filter
codaFS_selbal_wrapper(): Selbal-Style Forward Balance Feature Selection
codaFS_stability_logratio(): Stability Selection on Pairwise Log-Ratios
codacore-interface: CoDaCoRe Interfaces for Sparse Log-Contrast Classification
compare_gof_filters(): Compare GOF Filters on Task
compute_domain_shift(): Quantify Domain Shift Between Source and Target Platforms
compute_ece(): Compute Expected Calibration Error (ECE)
compute_nogueira_stability(): Compute Nogueira Stability Index
compute_shap_with_warnings(): Compute SHAP Values with Correlation Warnings
compute_stability_from_resample(): Compute Stability from ResampleResult
create_autoencoder_pipeop(): Create Autoencoder PipeOp
create_explainer(): Create Model Explainer
create_frozen_combat_pipeop(): Create a Frozen ComBat PipeOp for mlr3pipelines
create_hsfs_selector(): Create a Hybrid Sequential Feature Selector
create_mlp_learner(): Create MLP Learner via mlr3torch
create_omic_cache(): Create Split-Aware Cache
create_report_data(): Create Report Data Schema
create_stability_ensemble(): Create a Stability Ensemble with Presets
create_ws_pipeop(): Apply Within-Sample Normalization to OmicPipeline
cross-platform: Cross-Platform Transfer Learning Utilities
cross_platform_transfer(): Convenience Wrapper for Cross-Platform Transfer
dacvae_ct_to_abundance(): Convert qPCR Ct values to relative abundance with a fitted DA-cVAE's frozen reference
decompose_brier(): Decompose Brier Score
deploy_singlesample(): Build a frozen single-sample deployment scorer
encode_batch(): Batch-Encode an Expression Matrix with Any Encoding
encode_corr_grid(): Encode with Correlation-Ordered Grid
encode_deepinsight(): Encode with DeepInsight Layout
encode_ratio_image(): Create Pairwise Log-Ratio Image
encode_simple_grid(): Simple Row-Major Grid Encoding
evaluate_feature_selection(): Evaluate Feature Selection for Leakage
export_bundle(): Create Complete Export Bundle
export_mlr3torch_checkpoint(): Export mlr3torch Checkpoint
export_omicfit_checkpoint(): Export mlr3torch Checkpoint from OmicFit
export_onnx(): Export Model to ONNX Format
export_vetiver(): Export Model as Vetiver for Deployment
extract_features_from_resample(): Extract Features from All Folds in ResampleResult
extract_selected_features(): Extract Selected Features from Trained GraphLearner
feature_importance(): Compute Permutation Feature Importance
finetune_mlr3torch_checkpoint(): Fine-tune from mlr3torch Checkpoint
finetune_omicfit_checkpoint(): Fine-tune OmicFit from mlr3torch Checkpoint
fit_ai_scarf(): Fit the SCARF self-supervised contrastive single-sample discriminator
fit_bal_selbal(): Fit selbal-selected single-balance discriminator
fit_biofluid_anchor_rclr(): Fit biofluid-stable anchor rCLR
fit_biofluid_residualized_alr(): Fit biofluid-residualized ALR
fit_biofluid_stratified_rclr(): Fit biofluid-stratified rCLR centering
fit_block_fe_rclr(): Fit provenance-block rCLR centering moments
fit_coda_codacore(): Fit the CoDaCoRe stagewise log-ratio-balance discriminator
fit_coda_deepcoda(): Fit the DeepCoDA zero-sum log-contrast single-sample discriminator
fit_cohort_z_then_pool_score(): Fit cohort-wise robust z-score moments
fit_compositional_mahalanobis(): Fit robust Mahalanobis detector on compositional log-ratio coordinates
fit_conf_mondrian(): Fit Mondrian-conformal class-conditional LRT discriminator
fit_conformal_anomaly(): Fit a conformal anomaly detector from a Tier R healthy reference cohort
fit_corr_order(): Correlation-Ordered Grid Encoding
fit_cross_tech_harmonized_rclr(): Fit cross-technology harmonized rCLR anchors
fit_cvae(): Fit the counterfactual class-conditional VAE single-sample discriminator
fit_dacvae(): Fit the DA-cVAE single-sample discriminator
fit_dann(): Fit the DANN domain-adversarial cross-cohort single-sample discriminator
fit_deepinsight(): Fit DeepInsight Feature Layout
fit_dg_fishr(): Fit the Fishr cross-cohort transfer single-sample discriminator
fit_dg_ibirm(): Fit the IB-IRM cross-cohort transfer single-sample discriminator
fit_dominance_threshold(): Calibrate a cohort-specific dominance threshold
fit_dre_ulsif(): Fit unconstrained least-squares importance-fitting (uLSIF) density-ratio LRT
fit_dro_group(): Fit the Group-DRO kit x biofluid robust discriminator
fit_dro_vrex(): Fit the V-REx cross-cohort transfer single-sample discriminator
fit_ecod_copod(): Fit the ECOD + COPOD novelty discriminator
fit_frac_mfdfa(): Fit MFDFA spectrum (frac-mfdfa) within-sample discriminator
fit_frozen_quantile(): Fit a frozen monotone quantile calibrator from training data
fit_frozen_ruv(): Fit a frozen RUV factor model from training data
fit_group_dro_scorer(): Fit a Group-DRO kit x biofluid logistic scorer.
fit_gsp_gft(): Fit a graph-Fourier discriminator on a frozen co-expression graph
fit_hemolysis_kit_dual_anchor(): Fit the hemolysis plus kit-stable dual-anchor denominator.
fit_hemolysis_prefilter(): Fit a Hemolysis Pre-Filter
fit_hemolysis_rr(): Fit robust-regression hemolysis nuisance model from training controls
fit_icp(): Fit Invariant Causal Prediction discriminator (transfer at training, single-sample at inference)
fit_ig_fisherrao(): Fit Fisher-Rao geodesic class-conditional LRT discriminator
fit_img_gasfcnn(): Fit the GASF-image CNN single-sample discriminator
fit_inv_css(): Fit curvature-scale-space (inv-css) within-sample discriminator
fit_inv_fdaqf(): Fit quantile-function FDA (inv-fdaqf) within-sample discriminator
fit_inv_glcm(): Fit GLCM Haralick texture (inv-glcm) within-sample discriminator
fit_inv_olbp(): Fit ordinal Local Binary Pattern (inv-olbp) within-sample discriminator
fit_inv_scatter(): Fit the 1D wavelet-scattering single-sample discriminator
fit_isolation_forest_logratio(): Fit a pure-R isolation forest on rCLR log-ratio inputs
fit_isotonic_calibration(): Isotonic Regression Calibration
fit_kit_fe_adjusted_alr(): Fit kit fixed-effect adjusted ALR.
fit_kit_orthogonal_ilr(): Fit kit-orthogonal ILR-like residualization.
fit_kit_residual_mad(): Fit kit-residual MAD scoring.
fit_kit_stable_anchors(): Fit kit-stable denominator anchors on training data only.
fit_kit_stratified_rclr(): Fit kit-stratified rCLR centering sets.
fit_kme_witness(): Fit kernel mean-embedding witness-at-a-point discriminator
fit_logistic_normal_eb(): Fit frozen logistic-normal empirical-Bayes prior from training data
fit_lrt_bw(): Fit a Bures-Wasserstein Gaussian-OT class-conditional LRT discriminator
fit_lrt_copula(): Fit class-conditional Gaussian-copula LRT discriminator
fit_lrt_deepmaha(): Fit the deep-feature class-conditional Mahalanobis LRT discriminator
fit_lrt_nbkde(): Fit per-feature KDE naive-Bayes class-conditional LRT discriminator
fit_lrt_tcopula(): Fit class-conditional Student-t-copula LRT discriminator
fit_lrt_vinecopula(): Fit class-conditional vine-copula LRT discriminator
fit_man_nystrom(): Fit a Nystrom diffusion-map discriminator on frozen landmarks
fit_mixed_effects_scorer(): Fit disease weights from a cohort/provenance mixed-effects model
fit_moe_gated(): Fit the self-gated mixture-of-experts single-sample discriminator
fit_ot_lot(): Fit a linearized-optimal-transport (CDT tangent) LRT discriminator
fit_ot_slicedlrt(): Fit sliced random-projection Gaussian LRT discriminator
fit_platt_scaling(): Platt Scaling (Logistic Calibration)
fit_proto_net(): Fit the prototypical-network single-sample discriminator
fit_reo_ktsp(): Fit REO-kTSP within-sample pair-order discriminator
fit_reo_metaktsp(): Fit REO-MetaKTSP meta-analytic pair-order discriminator
fit_reo_pairratio(): Fit REO-pairratio within-sample log-ratio discriminator
fit_reo_rankboost(): Fit boosted rank trees external competitor
fit_reo_rankforest(): Fit a random rank forest external competitor
fit_reo_singscore(): Fit REO-singscore within-sample rank discriminator
fit_reo_ucell(): Fit REO-UCell within-sample rank discriminator
fit_rin_weighted_reference(): Fit a RIN-weighted frozen reference profile on training samples only.
fit_robust_pca_residual(): Fit a MAD-scaled SVD residual filter for batch correction
fit_sel_stablemate(): Fit StableMate stable-predictor discriminator (transfer at training, single-sample at inference)
fit_sig_path(): Fit the truncated path-signature single-sample discriminator
fit_singlesample_selector() print(<singlesample_selector_set>) print(<singlesample_selector_ineligible>): Fit a leakage-resistant single-sample method selector
fit_sinkhorn_ot_scorer(): Fit Sinkhorn OT cohort barycenter scorer.
fit_sinkhorn_single(): Fit the forced-single-sample entropic-OT (Sinkhorn) negative control
fit_ss_struct_ilr(): Fit the structurally-informed ILR single-sample scorer (ss-struct-ilr)
fit_ssl_vicreg(): Fit the VICReg self-supervised single-sample discriminator
fit_tabdpt(): Fit the TabDPT in-context discriminator (freeze the training context)
fit_tabicl(): Fit the TabICL in-context discriminator (freeze the training context)
fit_tabpfn(): Fit the TabPFN-v2 in-context discriminator (freeze the training context)
fit_tda_ph(): Fit topological persistence-image (tda-ph) within-sample discriminator
fit_tech_residualized_alr(): Fit technology-residualized ALR
fit_tech_stratified_rclr(): Fit technology-stratified rCLR centering moments
fit_temperature_scaling(): Temperature Scaling
fit_unc_sngp(): Fit the spectral-normalized neural Gaussian process discriminator (SNGP)
fit_ws_balance_ilr(): Fit within-cohort ILR balance discriminator
fit_ws_rclr_panel(): Fit a training-frozen trimmed-rCLR signed-panel score
frozen-combat: Frozen ComBat for Leakage-Free Batch Correction
frozen_combat_correct(): Convenience function for frozen ComBat correction
generate_cache_key(): Generate Split-Aware Cache Key
generate_gold_standard(): Generate Gold Standard Synthetic Dataset
generate_tripod_report(): Generate TRIPOD+AI Report
get_consensus_features(): Get Consensus Features from Best Signature
get_modality_info(): Get Modality Information
get_optimal_params(): Get Optimal Hyperparameters from AutoTuner
get_parallel_status(): Get Current Parallelization Status
get_reliable_shap_features(): Get Reliable SHAP Features
get_selected_features_per_fold(): Get Selected Features Per Fold
hemolysis-correction: Hemolysis-Aware Corrections for Biomarker Panels
hemolysis_index_blondal(): Blondal hemolysis index (log miR-451a - log miR-23a-3p)
hemolysis_proxy_score(): Hemolysis Proxy Score
image-encodings: Image Encoding Methods for Biomarker Panels
import_mlr3torch_checkpoint(): Import mlr3torch Checkpoint
import_omicfit_checkpoint(): Import mlr3torch Checkpoint into OmicFit
impute_within_fold(): Within-Fold Median Imputation (Leakage-Free)
interpretability: Model Interpretability for OmicSelector
is_singlesample_deployable(): Check whether a deployment scorer is single-sample deployable
load_bundle(): Load Exported Model Bundle
make_autotuner_glmnet(): Create Bayesian-Optimized AutoTuner for glmnet
make_autotuner_lightgbm(): Create Bayesian-Optimized AutoTuner for LightGBM
make_autotuner_ranger(): Create Bayesian-Optimized AutoTuner for Random Forest
make_autotuner_xgboost(): Create Bayesian-Optimized AutoTuner for XGBoost
make_catboost_learner(): Create a CatBoost Learner
make_fttransformer_learner() make_tabtransformer_learner(): Create an FT-Transformer Learner
make_gnn_learner(): Create GNN Learner for Pathway-Aware Classification
make_gof_filter(): Create GOF Filter
make_mlp_learner(): Make MLP Learner (Alias)
make_ratio_image(): Create Pairwise Ratio Image from Expression Vector
make_ratio_images(): Batch-Create Ratio Images from an Expression Matrix
make_tabm_learner(): Create a TabM Learner
make_tabnet_learner(): Create a TabNet Learner
make_tabpfn_learner(): Create a TabPFN Learner
memoize_with_split(): Memoize Function with Split Context
merge_omics_data(): Merge Multi-Omics Data for Analysis
methods-comparison: Methods Comparison Utilities
mimat_hsa_lookup: MIMAT accession \(\leftrightarrow\) hsa-miR name lookup
mirna_alias_table(): Curated miRNA alias lookup table (miRBase v22.1)
model-export: Model Export for OmicSelector
multi-omics: Multi-Omics Support for OmicSelector
noise_augment(): Gaussian Noise Augmentation
omic_benchmark(): Create benchmark service from OmicPipeline
omic_pipeline(): Quick pipeline creation from data
os_bias_audit(): One-Shot Bias Audit Report
os_bias_audit_report(): Plain-Text Bias-Audit Report
os_bias_floor_auc(): Dataset-Identity AUC ("Bias Floor")
os_calibrated_brier(): Compute Apparent or Cross-Fitted Calibrated Brier Score
os_claim_gate(): Claim Gate for Panel-Null Benchmarks
os_clustered_bootstrap_auc(): Clustered Bootstrap CI for AUC
os_conformal_anomaly(): Compute conformal anomaly p-value for new samples
os_conformal_anomaly_score(): Conformal Healthy-Reference Anomaly Score
os_covariate_only_auc(): Covariate-Only AUC
os_detect_cross_cohort_duplicates(): Specimen-Duplication Detection Across Cohorts
os_grouped_resample_auc(): Estimate AUC over Grouped Resampling Folds
os_identifiability_gate(): Apply a Fail-Closed Identifiability Gate
os_ktsp_fit(): Fit a Top-k Oriented-Pair Panel Scorer
os_log_transform_adaptive(): Adaptive Pseudocount log2 Transform
os_mahalanobis_score(): Mahalanobis Anomaly Score from a Reference Set
os_make_grouped_stratified_folds(): Build Grouped Stratified Folds
os_mde_margin_record(): Create an MDE/Margin Record
os_null_qc(): Validate a Random-Panel Null Benchmark
os_oof_pipeline_compare(): Out-of-Fold Pipeline Comparison
os_operating_point_gate(): Fail-Closed Operating-Point Gate
os_operating_points(): Compute Binary Clinical Operating Points
os_paired_delong(): Paired DeLong Test for AUC Comparison
os_panel_null_benchmark(): Matched Random-Panel Null Benchmark
os_per_feature_batch_signal(): Per-Feature Batch-vs-Case Signal Partitioning
os_plate_median_frozen: Frozen Plate-Median Correction
os_provenance_floor_suite(): Estimate a Provenance-Only Prediction Floor
os_provenance_preflight(): Single-sample specimen-overlap provenance pre-flight gate
os_required_margin(): Required Margin for Matched-Null Claims
os_singscore(): Score a Direction-Split Panel by Within-Sample Ranks
os_terminal_gate_ledger(): Build a Terminal Gate Ledger
os_validate_folds(): Validate Grouped Folds
os_within_provenance_blocks(): Summarize Within-Provenance Case-Control Blocks
parallel: Parallelization Support for OmicSelector
partial_dependence(): Compute Partial Dependence
plot_signature_tradeoffs(): Plot Signature Selection Trade-offs
plot_xai_importance(): Plot XAI Feature Importance
predict(<os_ktsp_model>): Predict from a Top-k Oriented-Pair Panel Scorer
predict_biofluid_anchor_rclr(): Predict biofluid-stable anchor rCLR scores
predict_biofluid_residualized_alr(): Predict biofluid-residualized ALR scores
predict_biofluid_stratified_rclr(): Predict biofluid-stratified rCLR scores
predict_block_fe_rclr(): Predict with a fitted block rCLR scorer
predict_clr_mlp(): Predict with a CLR + MLP Classifier
predict_codacore(): Predict with a CoDaCoRe Classifier
predict_cohort_z_then_pool_score(): Predict with a fitted cohort-z scorer
predict_cross_tech_harmonized_rclr(): Predict cross-technology harmonized rCLR scores
predict_mixed_effects_scorer(): Predict a mixed-effects panel score without reading test labels
predict_tech_residualized_alr(): Predict technology-residualized ALR scores
predict_tech_stratified_rclr(): Predict technology-stratified rCLR scores
predict_weighted_clr_mlp(): Predict with a Weighted CLR + MLP Classifier
preprocess_inverse_log(): Inverse-log preprocessor for pre-log-transformed microarray deposits
print(<CalibrationResult>): Print method for CalibrationResult
print(<CorrelationCheck>): Print Correlation Check
print(<FeatureImportance>): Print Feature Importance
print(<NestedCVResult>): Print method for NestedCVResult
print(<NogueiraStability>): Print method for NogueiraStability
print(<OmicBenchmarkResult>): Print method for OmicBenchmarkResult
print(<OmicFit>): Print method for OmicFit
print(<OmicsInput>): Print method for OmicsInput
print(<ReportData>): Print method for ReportData
print(<SignatureSelectionResult>): Print Method for Signature Selection Result
print(<singlesample_deployable>): Print a single-sample deployment object
print(<singlesample_selector>): Print a single-sample selector
print_shap_warnings(): Print SHAP Warnings Report
print_xai_summary(): Print XAI Summary
provenance_aware_scoring: Provenance-aware within-sample scoring methods
qpcr_nondetect_impute(): Bayesian hierarchical imputation of qPCR non-detects
qpcr_nondetect_lod_fallback(): Limit-of-detection fallback imputation for qPCR non-detects
ratio-image-cnn: Ratio Image CNN for Biomarker Panel Classification
register_coda_feature_selection_filters(): Register CoDA Feature Selection Filters
register_gof_filters(): Register GOF Filters in mlr3
reliability_diagram_data(): Create Reliability Diagram Data
report: TRIPOD+AI Report Generation for OmicSelector
reset_parallel(): Reset Parallelization to Sequential
resolve_hsa_to_mimat(): Resolve canonical hsa-miR-* names to MIMAT accession IDs
resolve_mimat_to_hsa(): Resolve MIMAT accession IDs to canonical hsa-miR-* names
resolve_mirna_aliases(): Resolve miRNA identifiers to a canonical namespace
run_bayesian_benchmark(): Run Bayesian Optimization Benchmark
run_dl_benchmark(): Deep Learning Benchmark
safe-evaluation: Safe Evaluation Utilities for Biomarker Validation
safe-preprocessing: Safe Preprocessing Utilities for Cross-Validation
safe_auc(): Safe AUC with Confidence Interval
safe_roc(): Safe ROC Computation (Direction-Guarded)
score_ai_scarf(): Score the SCARF self-supervised contrastive single-sample discriminator
score_bal_selbal(): Score selbal-selected single-balance discriminator
score_baseline_rclr(): Score baseline robust CLR for provenance-aware benchmark comparisons
score_biofluid_anchor_rclr(): Score biofluid-stable anchor rCLR
score_biofluid_residualized_alr(): Score biofluid-residualized ALR
score_biofluid_stratified_rclr(): Score biofluid-stratified rCLR
score_block_fe_rclr(): Score with provenance-block fixed-effect rCLR centering
score_cluster_robust_ensemble(): Score a cluster-robust ensemble of rCLR, ALR, and ILR-style components
score_coda_codacore(): Score the CoDaCoRe stagewise log-ratio-balance discriminator (pure base R)
score_coda_deepcoda(): Score the DeepCoDA zero-sum log-contrast single-sample discriminator
score_cohort_z_then_pool_score(): Score a panel after cohort-wise robust z-standardization
score_combat_seq_then_rclr(): Optional ComBat-seq followed by rCLR scoring.
score_conf_mondrian(): Score Mondrian-conformal class-conditional LRT discriminator
score_cross_tech_harmonized_rclr(): Score cross-technology harmonized rCLR
score_cvae(): Score the counterfactual class-conditional VAE single-sample discriminator
score_dacvae(): Score the DA-cVAE single-sample disease discriminator
score_dacvae_novelty(): Score the DA-cVAE single-sample novelty (out-of-distribution) capability
score_dann(): Score the DANN domain-adversarial cross-cohort single-sample discriminator
score_dg_fishr(): Score the Fishr cross-cohort transfer single-sample discriminator
score_dg_ibirm(): Score the IB-IRM cross-cohort transfer single-sample discriminator
score_dre_ulsif(): Score the uLSIF density-ratio LRT discriminator
score_dro_group(): Score the Group-DRO kit x biofluid robust discriminator
score_dro_vrex(): Score the V-REx cross-cohort transfer single-sample discriminator
score_ecod_copod(): Score the ECOD + COPOD novelty discriminator
score_frac_mfdfa(): Score MFDFA spectrum (frac-mfdfa) within-sample discriminator
score_group_dro_scorer(): Score held-out samples with a frozen Group-DRO scorer.
score_gsp_gft(): Score specimens with a frozen graph-Fourier discriminator
score_hemolysis_kit_dual_anchor(): Score a panel against the dual hemolysis plus kit-stable denominator.
score_icp(): Score Invariant Causal Prediction discriminator (single-sample)
score_ig_fisherrao(): Score Fisher-Rao geodesic class-conditional LRT discriminator
score_img_gasfcnn(): Score the GASF-image CNN single-sample discriminator (forced row-by-row)
score_inv_css(): Score curvature-scale-space (inv-css) within-sample discriminator
score_inv_fdaqf(): Score quantile-function FDA (inv-fdaqf) within-sample discriminator
score_inv_glcm(): Score GLCM Haralick texture (inv-glcm) within-sample discriminator
score_inv_olbp(): Score ordinal Local Binary Pattern (inv-olbp) within-sample discriminator
score_inv_scatter(): Score the 1D wavelet-scattering discriminator (forced row-by-row)
score_kit_fe_adjusted_alr(): Score kit fixed-effect adjusted ALR.
score_kit_orthogonal_ilr(): Score kit-orthogonal ILR-like residualized panel.
score_kit_residual_mad(): Score kit-residual MAD.
score_kit_stable_anchor_rclr(): Score a panel against train-only kit-stable anchors.
score_kit_stratified_rclr(): Score kit-stratified rCLR.
score_kme_witness(): Score kernel mean-embedding witness-at-a-point discriminator
score_lrt_bw(): Score a Bures-Wasserstein Gaussian-OT class-conditional LRT discriminator
score_lrt_copula(): Score class-conditional Gaussian-copula LRT discriminator
score_lrt_deepmaha(): Score the deep-feature class-conditional Mahalanobis LRT discriminator
score_lrt_nbkde(): Score per-feature KDE naive-Bayes class-conditional LRT discriminator
score_lrt_tcopula(): Score class-conditional Student-t-copula LRT discriminator
score_lrt_vinecopula(): Score class-conditional vine-copula LRT discriminator
score_man_nystrom(): Score specimens with a frozen Nystrom diffusion-map discriminator
score_mixed_effects_scorer(): Score with the mixed-effects scorer
score_moe_gated(): Score the self-gated mixture-of-experts single-sample discriminator
score_ot_lot(): Score a linearized-optimal-transport (CDT tangent) LRT discriminator
score_ot_slicedlrt(): Score sliced random-projection Gaussian LRT discriminator
score_platelet_exclusion_rclr(): Score rCLR after excluding blood-cell-derived miRNAs
score_proto_net(): Score the prototypical-network single-sample discriminator
score_rclr_baseline(): Baseline rCLR panel sum using the standard trimmed denominator.
score_reo_ktsp(): Score REO-kTSP within-sample pair-order votes
score_reo_metaktsp(): Score REO-MetaKTSP within-sample pair-order votes
score_reo_pairratio(): Score REO-pairratio within-sample log-ratio discriminator
score_reo_rankboost(): Score boosted rank trees one specimen at a time
score_reo_rankforest(): Score a random rank forest one specimen at a time
score_reo_singscore(): Score REO-singscore within-sample rank signature
score_reo_ucell(): Score REO-UCell within-sample rank signature
score_rin_weighted_score(): Score a panel against a RIN-weighted frozen training reference profile.
score_sel_stablemate(): Score StableMate stable-predictor discriminator (single-sample)
score_sig_path(): Score the truncated path-signature discriminator (pure base R, row-by-row)
score_singlesample_selector(): Score specimens with a frozen single-sample selector
score_sinkhorn_ot_scorer(): Score projected samples by summing panel features.
score_sinkhorn_single(): Score the forced-single-sample entropic-OT (Sinkhorn) negative control
score_specimen(): Score new specimens with a frozen single-sample deployment object
score_ss_struct_ilr(): Score the structurally-informed ILR single-sample scorer (ss-struct-ilr)
score_ssl_vicreg(): Score the VICReg self-supervised single-sample discriminator
score_tabdpt(): Score the TabDPT in-context discriminator (default row-by-row)
score_tabicl(): Score the TabICL in-context discriminator (default row-by-row)
score_tabpfn(): Score the TabPFN-v2 in-context discriminator (default row-by-row)
score_tda_ph(): Score topological persistence-image (tda-ph) within-sample discriminator
score_tech_residualized_alr(): Score technology-residualized ALR
score_tech_stratified_rclr(): Score technology-stratified rCLR
score_unc_sngp(): Score the spectral-normalized neural Gaussian process discriminator (SNGP)
score_ws_balance_ilr(): Score within-cohort ILR balance discriminator
score_ws_rclr_panel(): Score specimens with a frozen trimmed-rCLR signed panel
select_best_signature(): Select Best Biomarker Signature from Nested CV Results
select_cross_tech_anchor_features(): Select cross-technology anchor features on a training pool
setup_parallel(): Configure Parallelization for OmicSelector
shap_values(): Compute SHAP-like Values
shap_warnings: Correlation-Aware SHAP Interpretation
signature-selection: Signature Selection: Multi-Objective Best Biomarker Selection
singlesample-additional-within-sample: Single-sample additional within-sample methods (Module A, P2)
singlesample-batch-correction: Single-sample batch-correction methods (Module C)
singlesample-coda-codacore: CoDaCoRe stagewise log-ratio-balance single-sample discriminator (torch relaxation at fit, frozen discrete balances, PURE-R score)
singlesample-cvae: Counterfactual class-conditional VAE single-sample discriminator
singlesample-dacvae: Domain-adversarial conditional VAE single-sample discriminator (DA-cVAE)
singlesample-dann: DANN (domain-adversarial) cross-cohort transfer discriminator
singlesample-deepcoda: DeepCoDA zero-sum log-contrast bottleneck + self-explaining head (single-sample)
singlesample-deepmaha: Deep-feature class-conditional Mahalanobis LRT (learned frozen embedding)
singlesample-dg-fishr: Fishr (gradient-variance matching) cross-cohort transfer discriminator
singlesample-dg-ibirm: IB-IRM (Information-Bottleneck Invariant Risk Minimization) cross-cohort transfer discriminator
singlesample-dro-group: Group-DRO kit x biofluid robust discriminator (canonical single-sample wrapper)
singlesample-ecod-copod: ECOD + COPOD novelty discriminator (single-sample, frozen-ECDF)
singlesample-hemolysis: Single-sample robust-regression hemolysis correction (Module B)
singlesample-img-gasfcnn: GASF-image CNN single-sample discriminator (frozen-BatchNorm, python-at-score)
singlesample-inv-scatter: 1D wavelet-scattering single-sample discriminator (python-at-score, frozen head)
singlesample-matched-null: Single-sample matched-null benchmark for panel-vs-random-panel AUC inference
singlesample-moe-gated: Self-gated mixture of frozen experts (MoE-gated), single-sample
singlesample-nondetects: Single-sample qPCR non-detect imputation (Module B add-on)
singlesample-outlier-detection: Single-sample outlier detection and conformal claim-gating (Module D)
singlesample-preprocessing: Single-sample preprocessing utilities
singlesample-protonet: Prototypical-network single-sample discriminator (frozen class prototypes)
singlesample-scarf: SCARF self-supervised contrastive single-sample discriminator (frozen linear head)
singlesample-sig-path: Truncated path-signature single-sample discriminator (pure base-R, frozen head)
singlesample-sinkhorn-single: Forced-single-sample entropic-OT (Sinkhorn) negative control
singlesample-sngp: Spectral-normalized neural Gaussian process, distance-aware logit (SNGP)
singlesample-tabdpt: TabDPT in-context tabular-foundation discriminator (conditional single-sample)
singlesample-tabicl: TabICL in-context tabular-foundation discriminator (conditional single-sample)
singlesample-tabpfn: TabPFN-v2 in-context foundation-model discriminator (conditional single-sample)
singlesample-vicreg: VICReg self-supervised embedding + frozen linear-probe discriminator
singlesample-vrex: V-REx (Variance Risk Extrapolation) cross-cohort transfer discriminator
singlesample-within-sample: Single-sample within-sample compositional methods (Module A, P1)
singlesample_adjust_relevance_by(): Adjust the frozen directional relevance family with BY
singlesample_adjust_zero_by(): Adjust the frozen zero-difference test family with BY
os_fit_dann_kit_extended() os_score_dann_kit_extended() os_transform_dann_kit_extended() os_fit_kit_conditional_vae() os_score_kit_conditional_vae() os_transform_kit_conditional_vae() os_fit_icp_per_kit() os_score_icp_per_kit() os_predict_icp_per_kit_details() os_fit_group_dro_scorer() os_score_group_dro_scorer() os_fit_sinkhorn_ot_scorer() os_apply_sinkhorn_ot_scorer() os_score_sinkhorn_ot_scorer(): Advanced single-sample learned and transport scorers
singlesample_assert_method_bank_exports(): Assert that single-sample method-bank functions are present and exported
singlesample_assert_row_equivariant(): Assert that a scorer is row-equivariant (single-sample deployable)
singlesample_bh_fdr_correct_blocked(): Block-aware two-stage BH-FDR for specimen-shared cohort clusters
singlesample_bh_fdr_correct_matched_null(): BH-FDR correction within a matched-null modality family
singlesample_complete_support_panels(): Construct fixed eligibility-only complete-support panels
singlesample_corrected_repeated_cv(): Corrected repeated-CV inference for a paired method effect
singlesample_external_competitor_roster(): External rank-tree competitor registry
singlesample_hanley_mcneil_auc_ci(): Hanley-McNeil 1982 AUC confidence interval
singlesample_holm_correct_familywise(): Holm correction for family-wise method contrasts
singlesample_is_row_equivariant(): Test whether a scorer is row-equivariant (non-throwing)
singlesample_make_loco_splits(): Build leave-one-cohort-out splits with same-block exclusion
singlesample_make_locto_splits(): Build leave-one-cancer-type-out splits with same-block exclusion
singlesample_matched_null_benchmark(): Single-sample matched-null benchmark for within-sample miRNA panel scoring
singlesample_matched_null_benchmark_cv(): 5-fold nested-CV matched-null benchmark
singlesample_matched_pair_auc(): Matched AUC difference across repeated CV strata
singlesample_method_bank(): Single-sample method-bank registry
singlesample_method_roster(): Amendment #4 method roster (frozen expansion)
singlesample_paired_auc_diff_se(): Paired DeLong SE for single-sample AUC lift
singlesample_register_score_adapter(): Register a non-canonical score adapter for a roster method
singlesample_score_call(): Score a roster method with the canonical single-sample interface
singlesample_selector_candidates(): Eligible methods for the single-sample selector
singlesample_technology_lift_delong(): Paired DeLong SE for technology-aware transfer lift
singlesample_within_method_pairs(): Enumerate frozen within-method pairs
smote_augment(): SMOTE Augmentation for Omics Data
stability: Nogueira Stability Index for Feature Selection
stack_omics(): Create Multi-Omics Stacked Ensemble (Convenience Function)
standardize_within_fold(): Within-Fold Standardization (Leakage-Free)
tabddpm_generate(): TabDDPM Synthetic Data Generator
technology_aware_scoring: Technology-aware within-sample scoring methods
test_noninferiority(): Paired DeLong Non-Inferiority Test
torch-checkpoint: mlr3torch Checkpoint Utilities
torch-learners: mlr3torch Learner Integration for OmicSelector
train_clr_mlp(): Train a CLR + MLP Classifier
train_codacore(): Train a CoDaCoRe Classifier
train_ratio_cnn(): Train Ratio Image CNN
train_ratio_cnn_multiseed(): Train Ratio CNN Across Three Default Seeds
train_weighted_clr_mlp(): Train a Weighted CLR + MLP Classifier
validate_omics_input(): Validate Multi-Omics Input
validate_synthetic(): Validate Synthetic Data Quality
weighted_clr_transform(): Weighted Centered Log-Ratio Transformation
with_parallel(): With Parallel Scope
within-sample: Within-Sample Normalization for Biomarker Panels
ws_alr_pivot(): Additive log-ratio with a frozen pivot pool
ws_balance_ilr(): Isometric log-ratio balances on a frozen partition tree
ws_default_pivot_pool(): Default ALR pivot pool for circulating-miRNA panels (v1)
ws_default_sbp(): Default circulating-miRNA sequential binary partition (v1)
ws_dominance_flag(): Flag samples whose dominance score exceeds a calibrated threshold
ws_dominance_score(): Panel-internal dominance score (within-sample QC)
ws_logratio(): Within-Sample Pairwise Log-Ratio Features
ws_mad_logratio(): Median-centred log-ratio with optional MAD scaling
ws_minmax(): Within-Sample Min-Max Normalization
ws_perturbation_benchmark(): Within-Sample Perturbation Benchmark
ws_rank(): Within-Sample Rank Normalization
ws_ratio_image(): Within-Sample Pairwise Ratio Image
ws_rclr_trimmed(): Robust trimmed centred log-ratio for compositional miRNA panels
ws_zscore(): Within-Sample Z-Score Normalization
xai_correlations(): Compute Correlation Diagnostics for Features
xai_explainer_mlr3(): Create DALEX Explainer from mlr3 Learner
xai_importance(): Compute Permutation Feature Importance
xai_pdp(): Compute Partial Dependence Plots
xai_pipeline(): Run Complete XAI Pipeline
xai_shap(): Compute SHAP Values for Observations