Package: pdactrace 0.99.25

Jibeom Ko

pdactrace: Stage-Aware PDAC Atlas and Prospectively Frozen Tissue-to-Serum Biomarker Audit Framework

pdactrace implements a transparent, deterministic claim-audit framework for stage-aware pancreatic ductal adenocarcinoma (PDAC) tissue-to-blood biomarker evidence. It integrates per-stage trajectory matching across bulk RNA-seq, tissue proteomics, single-cell origin, serum proteomics, and pancreatitis context, with explicit anchor-enrichment evaluation against a curated external reference set. The framework comprises a 12-template competitive trajectory catalog, a 3-axis + 2-gate reference score, an Evidence Math layer, TRACE-D tissue-to-serum directional evidence concordance, claim-tier classification, and weight-robustness and Pareto support audits. A prespecification layer supports freezing candidate scores, directions and a decision rule under a content hash before outcome labels are opened, with a distinct inconclusive terminal state when a frozen gate fails. Scope is deliberately PDAC-specific: the bundled atlas, the anchor reference set and every validated claim are PDAC, and no result is claimed for any other disease. Separately, no supervised classifier is trained or shipped, because validated non-circular early-detection ground truth is unavailable in PDAC.

Authors:Jibeom Ko [aut, cre]

pdactrace_0.99.25.tar.gz
pdactrace_0.99.25.zip(r-4.7-any)pdactrace_0.99.25.zip(r-4.6-any)pdactrace_0.99.25.zip(r-4.5-any)
pdactrace_0.99.25.tgz(r-4.6-any)pdactrace_0.99.25.tgz(r-4.5-any)
pdactrace_0.99.25.tar.gz(r-4.7-any)pdactrace_0.99.25.tar.gz(r-4.6-any)
pdactrace_0.99.25.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
pdactrace/json (API)

# Install 'pdactrace' in R:
install.packages('pdactrace', repos = c('https://biocstaging.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/jibeomko/pdactrace/issues

Datasets:

On CRAN:

Conda:

softwarebiomedicalinformaticsclassificationtranscriptomicsproteomicsgeneexpressiondifferentialexpressionsinglecellvisualizationannotation

4.20 score 92 exports 36 dependencies

Last updated from:131aa2c00e. Checks:1 NOTE, 9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
bioc-checksNOTE284
linux-develOK424
source / vignettesOK477
linux-releaseOK459
macos-releaseOK423
macos-oldrelOK385
windows-develOK446
windows-releaseOK405
windows-oldrelOK344
wasm-releaseOK165

Exports:align_patientalign_patient_profileanchor_enrichmentas_summarized_experimentassemble_user_evidenceatlas_provenancebuild_evidence_graphcase_studyclassify_claim_tierclassify_prot_trajectoryclassify_protein_trajectoryclassify_trajectorycompare_candidatescompare_genescompute_audit_scorecompute_pareto_classcompute_pareto_layerscompute_trace_ddownload_phase_csvsearly_pattern_namesearly_patternsevaluate_anchor_enrichmentevaluate_pareto_stabilityevidence_mathexplain_geneexplain_scoreexplain_user_evidence_modelextract_graph_featuresfit_stage_defit_stage_de_proteinfit_user_evidence_modelformat_provenancelist_atlas_metadatalist_candidateslist_data_sourcesmake_evidence_featuresmid_pattern_names_excludedmid_patternsmodel_cardNCS_W_15COLNCS_W_DOUBLENCS_W_SINGLENCS_W_TRIPLEpdactrace_axes_themepdactrace_pal_classpdactrace_pal_dirpdactrace_pal_grouppdactrace_pal_patternpdactrace_panel_themepdactrace_saveplot_candidate_landscapeplot_celltype_fullplot_filter_diagnosticsplot_filter_traceplot_gene_evidenceplot_gene_hexagonplot_gene_templateplot_meta_forestplot_panel_heatmapplot_per_cohortplot_serum_directionplot_stage_effectplot_template_atlaspredict_user_evidence_modelproject_user_cohortproject_user_serum_cohortpropagate_uncertaintyquery_genequery_gene_detailedquery_panelreport_generun_weight_robustnesssb_average_precisionsb_binary_aucsb_default_one_shot_rulesb_loco_stabilitysb_method_scoressb_multiblock_plsda_scoressb_observabilitysb_one_shotsb_pareto_frontssb_simulatesb_topk_recallsb_validate_effectsschema_specscore_anchor_similarityscore_trajectorystagebridgesummarize_gene_evidencesummarize_translation_gaptrace_filtersviz_gene

Dependencies:abindBiobaseBiocGenericsclicpp11data.tableDelayedArrayfarvergenericsGenomicRangesggplot2gluegtableIRangesisobandlabelinglatticelifecycleMatrixMatrixGenericsmatrixStatspatchworkR6RColorBrewerrlangS4ArraysS4VectorsS7scalesSeqinfoSparseArraySummarizedExperimentvctrsviridisLitewithrXVector

Methodology validation: cutoff sensitivity + tool comparison
Goal | Section A -- Cutoff sensitivity | A1. Trajectory classification cutoff (rho_cutoff) | A2. Three-axis weight sensitivity | Section B -- Tool comparison | Section C -- Negative control depletion | C1. Top-N depletion (negative controls absent from the most-prioritised candidates) | C2. Bottom-N enrichment (negative controls concentrated at the lowest-priority end) | Section D -- Held-out PDAC cohort projection (v0.99.12) | Section E -- Three-way baseline: where does enrichment come from? | Limitations of these analyses
Rendered frommethodology_validation.Rmdusingknitr::rmarkdown

Last update: 2026-09-28
Started: 2026-05-10

Audit framework: 3 axes + 2 gates + 4 classes (v0.3.0)
Goal | 1. Why a hand-engineered rule (and not a supervised model) | 2. The 6-function API at a glance | 3. LTBP1 — the supported_uncertain exemplar | 4. GAPDH — the housekeeping negative control | 5. Anchor enrichment evaluation (post-freeze) | 6. Reading the case_study() output | 7. What the audit framework does not claim | Session info
Rendered fromaudit_framework.Rmdusingknitr::rmarkdown

Last update: 2026-09-28
Started: 2026-05-07

Prespecified scoring and the frozen one-shot audit
Goal | Step 1 -- A simulated multi-cohort, multi-omic benchmark | Step 2 -- Depth-adjusted blood observability | Step 3 -- Rank candidates | Step 4 -- Does the ranking survive dropping a cohort? | Step 5 -- Comparators, scored honestly | Step 6 -- The frozen one-shot audit | Why a calibration check matters | Scope
Rendered fromprespecification_audit.Rmdusingknitr::rmarkdown

Last update: 2026-09-28
Started: 2026-09-27

Reproducibility from raw data
Goal | Layer 1 — Offline, bundled | Layer 2 — User cohort | Layer 3 — Re-derive the bundled atlas | Layer 4 — From FASTQ | Provenance summary | Session information
Rendered fromreproducibility.Rmdusingknitr::rmarkdown

Last update: 2026-09-27
Started: 2026-05-08

Audit-class case studies — LGALS3BP / LTBP1 / ALB / GAPDH
Goal | Side-by-side audit summary | Detailed audit features per gene | Case 1 — LGALS3BP (high_confidence anchor) | Case 2 — LTBP1 (supported_uncertain, Class B inverse exemplar) | Case 3 — ALB (penalized, plasma high-abundance gate) | Case 4 — GAPDH (excluded, housekeeping gate) | Take-away | Session information
Rendered fromaudit_case_studies.Rmdusingknitr::rmarkdown

Last update: 2026-09-27
Started: 2026-05-07

Apply pdactrace to your own multi-omics cohort
Goal | Step 1 — Load the bundled toy cohort | Step 2 — Fit the stage-aware RNA model | Step 3 — Fit the protein limma model | Step 4 — Assemble the user evidence table | Step 5 — Run the audit framework on user evidence | Step 6 — Bioconductor-native input via SummarizedExperiment (v0.99.2) | Step 7 — Single-patient trajectory alignment (v0.4.1) | Scope boundary | Appendix A — Stage harmonization (TNM / AJCC -> Normal / Early / Mid / Late) | Appendix B -- Serum projection (binary group contrast, no stage)
Rendered fromuser_cohort_extension.Rmdusingknitr::rmarkdown

Last update: 2026-09-27
Started: 2026-05-07

Portability of the R API to a non-PDAC cohort
Goal | Step 1 -- Build a synthetic non-PDAC cohort | Step 2 -- Project through the framework (one call) | Step 3 -- What changes between PDAC and non-PDAC? | Step 4 -- Translation discipline transfers, but the join target changes | Limitations
Rendered fromcross_cancer_demonstration.Rmdusingknitr::rmarkdown

Last update: 2026-09-27
Started: 2026-05-10

Lookup basics: querying the pdactrace atlas
Goal | 1. Single-gene query — query_gene() | 2. Multi-gene comparison — query_panel() | 3. Multi-parameter candidate listing — list_candidates() | 4. Filter audit trail — trace_filters() | 5. Plain-text summary — summarize_gene_evidence() | Atlas metadata | Next | Session
Rendered fromlookup_basics.Rmdusingknitr::rmarkdown

Last update: 2026-05-27
Started: 2026-05-07

Readme and manuals

Help Manual

Help pageTopics
pdactrace . PDAC-TRACE: queryable stage-aware tissue-to-serum biomarker atlaspdactrace-package pdactrace
Align a single patient's omics profile against atlas stage trajectoriesalign_patient align_patient_profile
Convert pdactrace_reference into a SummarizedExperimentas_summarized_experiment
Assemble user multi-omics inputs into a per-gene evidence tableassemble_user_evidence
Atlas metadataatlas_metadata
Atlas provenance (version, build date, source repo, DOIs)atlas_provenance
Build a transparent evidence graph for one genebuild_evidence_graph
Summarize audit-score case-study genescase_study
Classify tissue-to-blood biomarker claim tiersclassify_claim_tier
Classify each protein's trajectory against the 12-template catalogclassify_protein_trajectory classify_prot_trajectory
Classify each gene's trajectory against the 12-template catalogclassify_trajectory
Compare multiple candidate genes side-by-sidecompare_candidates
Side-by-side Evidence Math comparison across multiple genescompare_genes
Compute the frozen pdactrace audit score (3-axis + 2-gate)compute_audit_score
Classify Pareto evidence supportcompute_pareto_class
Compute weight-free Pareto layers on the three evidence axescompute_pareto_layers
Algorithm 1 — TRACE-D: Tissue-to-serum Directional Evidence Concordancecompute_trace_d
Default trajectory templates (12-template catalog, v0.4.0)default_templates
Download the two large upstream phase CSVs into a local cachedownload_phase_csvs
Early-onset pattern names (the 4 atlas-surfaced patterns)early_patterns early_pattern_names
Evaluate external-anchor enrichment in top-ranked genesanchor_enrichment evaluate_anchor_enrichment
Evaluate Monte-Carlo Pareto stability under perturbed evidenceevaluate_pareto_stability
Per-axis mathematical evidence for one geneevidence_math
Print a per-axis evidence summary for one geneexplain_gene
Decompose a gene's audit score into its three axes and two gatesexplain_score
Explain a 'pdactrace_user_model' via its top contributing featuresexplain_user_evidence_model
Extract graph-aware audit featuresextract_graph_features
Fit a stage-aware DESeq2 LRT model on user RNA-seq datafit_stage_de fit_stage_de,ANY-method fit_stage_de,SummarizedExperiment-method
Fit a stage-aware limma model on user tissue-protein intensity datafit_stage_de_protein fit_stage_de_protein,ANY-method fit_stage_de_protein,SummarizedExperiment-method
Fit a user-supplied supervised model on the evidence feature matrixfit_user_evidence_model
Human-readable evidence labels for the bundled phase tagsformat_provenance
Inspect atlas metadatalist_atlas_metadata
List candidate genes by biology-level filter parameterslist_candidates
List the public datasets that contributed to the pdactrace atlaslist_data_sources
Build a flat numeric feature matrix for downstream modellingmake_evidence_features
Random-effects meta-analysis tablemeta_analysis
Mid-onset pattern names (excluded from atlas surface)mid_patterns mid_pattern_names_excluded
Model card metadata for the optional ML layermodel_card
Public-data accession table backing list_data_sources()pdactrace_data_sources
External PDAC biomarker anchor provenance tablepdactrace_external_anchors
Per-stage tissue-protein effect-size table for trajectory plottingpdactrace_protein_betas
Save a ggplot to PDF using cairo_pdfpdactrace_save
Palettes for pdactrace figurespdactrace-palettes pdactrace_pal_class pdactrace_pal_dir pdactrace_pal_group pdactrace_pal_pattern
pdactrace NCS-grade ggplot2 themeNCS_W_15COL NCS_W_DOUBLE NCS_W_SINGLE NCS_W_TRIPLE pdactrace-theme pdactrace_axes_theme pdactrace_panel_theme
Visualise the candidate landscape across tissue and serumplot_candidate_landscape
Full 11-celltype expression barplot_celltype_full
Filter diagnostic step bar with underlying metric annotationplot_filter_diagnostics
Visualise tissue-to-serum filter passage with serum direction contextplot_filter_trace
Multi-panel ggplot composite of gene evidenceplot_gene_evidence
Hexagonal evidence radar for one or more genesplot_gene_hexagon
Per-gene template-overlay plotplot_gene_template
Random-effects meta-analysis forest plot per geneplot_meta_forest
Compare a gene panel across evidence layers as a heatmapplot_panel_heatmap
Per-cohort sign vote bar with Stouffer summaryplot_per_cohort
Per-gene serum direction context, coloured by translation classplot_serum_direction
Forest plot of per-stage effect sizeplot_stage_effect
Per-template trajectory atlas plotsplot_template_atlas
Predict from a 'pdactrace_user_model'predict_user_evidence_model
End-to-end user-cohort projection through the audit frameworkproject_user_cohort
Project a user-supplied serum proteomics cohort (binary contrast)project_user_serum_cohort
Propagate audit-score uncertaintypropagate_uncertainty
Query the pdactrace reference atlas for a single genequery_gene
Detailed gene-level evidence - 5 granularity tablesquery_gene_detailed
Query the pdactrace reference atlas for multiple genesquery_panel
Render a pdactrace evidence report for one gene or a panelreport_gene
Evaluate ranking robustness across plausible audit weightsrun_weight_robustness
Average precisionsb_average_precision
Binary area under the ROC curvesb_binary_auc
Default frozen thresholds for 'sb_one_shot()'sb_default_one_shot_rule
Leave-one-cohort-out rank stabilitysb_loco_stability
Score candidates with StageBridge and benchmark comparatorssb_method_scores
Score shared features with matched-sample multiblock PLS-DAsb_multiblock_plsda_scores
Calculate depth-adjusted blood observabilitysb_observability
Run one prospectively locked destination-cohort audit to a terminal decisionsb_one_shot
Assign exact non-dominated Pareto frontssb_pareto_fronts
Simulate a multi-cohort, multi-omic biomarker benchmarksb_simulate
Recall among the top-ranked candidatessb_topk_recall
Validate a StageBridge effect tablesb_validate_effects
Schema column specificationschema_spec
Descriptive feature-space similarity to bundled external anchorsscore_anchor_similarity
Score one or more genes against all 12 templatesscore_trajectory
Rank stage-resolved multi-omic biomarker candidatesstagebridge
Human-readable text summary of gene evidencesummarize_gene_evidence
Summarize the tissue-to-blood translation gapsummarize_translation_gap
Synthetic stage / cohort metadata for runnable examplestoy_coldata
Synthetic RNA count matrix for runnable examplestoy_counts
Synthetic tissue-protein log2 intensity matrix for runnable examplestoy_protein
Trace which selection filters a gene passed or failedtrace_filters
Single-call visual evidence + scoring canvas for one geneviz_gene