Function reference

Use the workflow chapters for complete analysis delivery. This map preserves the detailed public function guide for direct R lookup. Open ?function_name in R for the exact interface in your installed version.

How To Use This Page

This map is organized by user goal, not by source file. Internal helpers are not included. Start with the goal that matches your task.

Data Preparation

Function	Use when	Output	New user?	Boundary
`iobr_causal_load()`	load clinical/signature data into CausalData	CausalData	yes	data prep only
`causal_from_iobr()`	bridge IOBR objects/results	CausalData-like object	yes	bridge, not evidence
`iobr_causal_impute()`	impute missing values in CausalData	imputed CausalData	cautious	imputation adds assumptions

Main Workflow

Function	Use when	Output	New user?	Boundary
`iobr_causal_pathway_first_workflow()`	expression plus clinical data; default scientific main route	pathway and gene endpoint reports	yes, start here	pathway evidence, not mechanism proof
`iobr_causal_gene_set_sources()`	none	supported source/license table	before external retrieval	availability and license boundary only
`iobr_causal_get_gene_sets()`	source plus optional pathway names	named gene sets or audited resource	pathway-first input preparation	database content, not pathway evidence
`iobr_causal_match_gene_sets()`	pathway names plus gene sets	conservative match audit	reconcile score and gene-set names	no unrestricted fuzzy matching
`iobr_causal_map_gene_ids()`	human gene identifiers	mapping audit table	identifier preparation	mapping is not expression QC
`iobr_causal_pathway_first_rules()`	configure pathway and mapped-gene gates	evidence rules	yes	user-configurable prioritization rules
`iobr_causal_validate_main_route()`	verify known-truth recovery, parameter reach, parallel parity, and reference agreement	validation tables	maintainer/reviewer	software QA, not external biological validation
`iobr_causal()`	run a general high-level workflow	workflow result	secondary	still assumption-bound
`causal_pipeline()`	flexible causal pipeline	pipeline result	intermediate	not automatic proof
`iobr_causal_exposure_first_workflow()`	prespecified/precomputed feature table	exposure-level reports	secondary	no pathway-to-gene discovery

Pathway-first is the organizing main line. Candidate screening, formal-method extensions, survival special cases, prediction, visualization, AI, and audit support that line or answer narrower questions; they do not replace it.

Parameter Control

Function	Use when	Output	New user?	Boundary
`iobr_causal_control()`	set thresholds/methods in one object	control object	yes	parameters must be reported
`iobr_causal_control_reference()`	explain parameters	data frame	yes	explanation only

Secondary Candidate Screening

Function	Use when	Output	New user?	Boundary
`iobr_causal_define_exposure_pool()`	define candidate features	exposure pool	intermediate	not evidence
`iobr_causal_prefilter_exposures()`	remove unusable candidates	prefilter result	intermediate	quality control
`iobr_causal_scan_exposures()`	scan candidates against endpoint	scan table	intermediate	screening only
`iobr_causal_candidate_gate()`	select candidates for formal review	gate table	yes	gate, not conclusion
`iobr_causal_intersect_exposures()`	find ORR/survival overlap	intersection table	yes	priority only
`iobr_causal_candidate_workflow()`	run staged workflow on prespecified features	workflow + Excel outputs	secondary	candidate entry, not pathway-first replacement
`iobr_causal_longitudinal_candidate_scan()`	scan longitudinal candidates	scan result	advanced	exploratory unless supported

Formal Estimation

Function	Use when	Output	New user?	Boundary
`iobr_causal_estimate()`	estimate effect from CausalData	estimate result	yes	under declared assumptions
`iobr_causal_formal_estimate_candidates()`	estimate selected candidates	formal candidate table	yes	not multi-method proof by default
`causal_glm()`	GLM estimate	estimate table	yes	model-based
`causal_gcomp()`	g-computation	estimate table	yes	model-based standardization
`causal_ipw()`	inverse probability weighting	estimate table	advanced	check weights/ESS
`causal_psm()`	propensity score matching	matched estimate	advanced	check match quality
`causal_tmle()`	TMLE	estimate table	advanced	dependency/assumption sensitive
`causal_aipw()`	doubly robust AIPW	estimate table	advanced	needs diagnostics
`causal_dml()`	double machine learning	estimate table	advanced	needs careful review
`causal_gps()`	continuous exposure GPS	estimate table	advanced	continuous exposure assumptions
`causal_forest_estimate()`	heterogeneity ranking	HTE result	advanced	not individualized proof
`causal_compare_all()`	compare methods	method comparison	intermediate	comparison, not consensus proof
`compare_methods()`	summarize method results	table	intermediate	display utility

Feature, Discovery, DAG

Function	Use when	Output	New user?	Boundary
`iobr_causal_feature()`	select features	selected features	intermediate	feature selection only
`causal_select()`	single feature selection method	selection result	intermediate	screening
`causal_select_ensemble()`	ensemble feature selection	selection result	advanced	screening
`iobr_causal_discover()`	DAG/causal discovery	discovery result	advanced	diagnostic only
`causal_discover_single()`	one discovery method	graph result	advanced	not proof of identification
`causal_discover_ensemble()`	ensemble discovery	graph result	advanced	support, not proof
`iobr_causal_validate_adjustment()`	review adjustment set	validation result	advanced	adjustment-boundary review
`causal_check_adjustment()`	check DAG adjustment	validation result	advanced	DAG-dependent
`dag_evaluate()`	evaluate discovery output	diagnostic table	advanced	diagnostic

Survival

Function	Use when	Output	New user?	Boundary
`iobr_causal_survival_report()`	standard survival report	survival report	yes	check PH/censoring
`iobr_causal_survival_platform_report()`	unified survival report	platform report	yes	stable/extended scopes differ
`iobr_causal_survival_diagnostics()`	survival diagnostics	diagnostic table	yes	diagnostic
`survival_diagnostics()`	time-to-event diagnostics	diagnostic table	yes	diagnostic
`iobr_causal_survival_rmst()`	RMST summary	RMST table	intermediate	horizon-specific
`iobr_causal_survival_milestone()`	milestone survival	milestone table	intermediate	timepoint-specific
`iobr_causal_survival_competing_risks()`	competing risk summary	competing risk result	advanced	competing event assumptions
`iobr_causal_survival_competing_report()`	competing risk report	report	advanced	bounded
`iobr_causal_survival_vnext_report()`	extended survival family report	vNext report	advanced	research/extended
`iobr_causal_survival_unified_report()`	alias for unified report	report	yes	same as platform
`iobr_causal_survival_final_summary()`	alias for final report	report	yes	compatibility
`iobr_causal_survival_maximal_report()`	maximal survival report alias	report	advanced	compatibility

Advanced Survival Families

Family	Functions	Boundary
longitudinal survival	`iobr_causal_survival_longitudinal_prepare()`, `iobr_causal_survival_msm()`, `iobr_causal_survival_gformula()`, `iobr_causal_survival_regime_compare()`, `iobr_causal_survival_regime_engine()`	advanced, assumption-heavy
dynamic regimes	`iobr_causal_survival_dynamic_regimes_prepare()`, `iobr_causal_survival_dynamic_regime_compare()`	prototype/policy review unless supported
longitudinal competing risks	`iobr_causal_survival_longitudinal_competing_prepare()`, `iobr_causal_survival_longitudinal_competing_msm()`, `iobr_causal_survival_longitudinal_competing_gformula()`, `iobr_causal_survival_longitudinal_competing_report()`	advanced
multistate	`iobr_causal_survival_multistate_prepare()`, `iobr_causal_survival_multistate_summary()`, `iobr_causal_survival_multistate_causal_prepare()`, `iobr_causal_survival_multistate_causal_estimate()`, `iobr_causal_survival_multistate_report()`	semi-causal/advanced
policy learning	`iobr_causal_survival_policy_prepare()`, `iobr_causal_survival_policy_learn()`, `iobr_causal_survival_policy_evaluate()`, `iobr_causal_survival_policy_report()`	prototype/exploratory unless support says otherwise
time-local	`iobr_causal_survival_time_local()`	local time summary

Sensitivity And Diagnostics

Function	Use when	Output	Boundary
`iobr_causal_sensitivity()`	robustness supplement	sensitivity result	not proof of no confounding
`causal_ps_check()`	propensity/balance check	diagnostics	support review
`causal_evaluate()`	conservative result evaluation	evaluation	not a new estimate
`causal_summary()`	summarize estimates	summary	display

Interpretation

Function	Use when	Output
`iobr_causal_readout()`	user-friendly result readout	text/list readout
`iobr_causal_contract()`	normalize result contract	contract object
`iobr_causal_interpret()`	unified interpretation helper	interpretation
`interpret_estimate()`	estimate-specific interpretation	interpretation
`interpret_survival()`	survival-specific interpretation	interpretation
`interpret_discovery()`	discovery/DAG interpretation	interpretation
`interpret_validation()`	validation interpretation	interpretation
`interpret_hte()`	heterogeneity interpretation	interpretation

Excel, Artifacts, Audit

Function	Use when	Output
`iobr_causal_read_result_bundle()`	read Excel/result directory	bounded readout
`iobr_causal_write_audit_summary()`	summarize an existing audit folder	audit summary
`iobr_causal_artifacts()`	list artifacts	artifact table
`iobr_causal_audit()`	audit claim/method/rerun surfaces	audit table
`iobr_causal_method_provenance()`	explain method source	provenance table
`iobr_causal_freeze_protocol()`	freeze protocol	protocol file/list
`iobr_causal_rerun_recipe()`	show rerun recipe	recipe
`iobr_causal_log()`	view/export log	log table/file
`iobr_causal_selftest()`	package self-test	test results

Visualization

Function	Use when
`iobr_causal_plot()`	one contract-aware plot
`iobr_causal_visual_report()`	ordered figure set
`plot_clinical_summary()`	clinical-style summary figure
`plot_clinical_implications()`	clinical implication summary

Ordinary users should choose plot surfaces through iobr_causal_plot(type = ...). For example, use type = "forest", "love", "screening", "sensitivity", or "survival". Lower-level plot helpers are implementation details unless they are exported in the installed namespace.

AI and Agent Entry

Function	Use when
`iobrcausal()`	structured requests, configured natural language, or interactive R

CLI/MCP use the same request dispatcher. Provider and registry helpers are internal.

Platform Design Families

Function	Use when	Boundary
`iobr_causal_iv_report()`	instrumental-variable/LATE design	IV assumptions required
`iobr_causal_did_report()`	difference-in-differences design	parallel trends review
`iobr_causal_rdd_report()`	regression discontinuity design	local cutoff/bandwidth review

These are design-family reports. Do not describe them as ordinary adjusted observational estimates.