UKBAnalytica

UKBAnalytica is an R package for analyzing with the UK Biobank Research Analysis Platform (RAP) data inside approved RAP projects. It focuses on standardized phenotyping, survival-ready datasets, scalable preprocessing, and downstream analysis.

The package does not ship UK Biobank participant-level source records. Examples use field IDs, simulated toy data, or user-provided tables that remain within RAP-controlled storage.

Get started View GitHub

What the package covers

RAP-native extraction

Plan and extract UK Biobank fields within the approved RAP environment.

Baseline preprocessing

Standardize common baseline variables, recode UKB missing-value conventions, and derive reusable covariates.

Multi-source phenotyping

Define disease endpoints using ICD-10, ICD-9, self-report, death registry, OPCS4, cancer registry, first-occurrence, and algorithm-defined outcomes.

Survival-ready cohorts

Separate prevalent history from incident outcomes and compute follow-up time for downstream Cox and survival analyses.

Downstream analysis

Run baseline tables, regression workflows, imputation, subgroup analysis, propensity score methods, mediation, and sensitivity analyses.

ML and omics modules

Support machine-learning workflows, SHAP interpretation, and bioinformatics analyses like proteomics enrichment, and STRING-based PPI analysis.

Workflow overview

Overview of the UKBAnalytica workflow

Installation

# install.packages("devtools")
devtools::install_github("Hinna0818/UKBAnalytica")

# use pkg like:
library(UKBAnalytica)

Minimal example

Run analyses that touch real UK Biobank data inside RAP only.

library(UKBAnalytica)
library(data.table)

## suggest you have a data.frame on your RAP 
ukb_data <- fread("population.csv")

## label diseases using the predefined ID in this pkg
diseases <- get_predefined_diseases()[
  c("AA", "Hypertension", "Diabetes")
]

## build survival dataset for downstream analyses
analysis_dt <- build_survival_dataset(
  dt = ukb_data,
  disease_definitions = diseases,
  prevalent_sources = c("ICD10", "ICD9", "Self-report", "Death"),
  outcome_sources = c("ICD10", "ICD9", "Death"),
  primary_disease = "AA",
  show_flow = TRUE,
  dt_threads = 8
)

Privacy boundary

UKBAnalytica documentation and skills are designed for RAP-based analysis. Participant-level UK Biobank data should remain inside approved RAP project storage. For AI-assisted scripting, users should describe the dataframe schema and intended analysis only; synthetic toy data can be used for interface tests, while real analyses are run by the user inside RAP. Only aggregate outputs or non-identifying rendered figures should be shared for interpretation.

About

Author

UKBAnalytica is developed and maintained by Nan He (Hinna) from the Department of Bioinformatics, School of Basic Medical Sciences, Southern Medical University.

Contact Maintainer

Citation

If you use UKBAnalytica in your study, please cite:

DOI pending