DROMA.Meta: Application Package Built on DROMA.R

Overview

DROMA.Meta is an R application package built on top of DROMA.R and DROMA.Set.
Its role is not to replace the core analysis packages, but to package a concrete multi-step biomarker discovery workflow that combines:

1. preclinical batch screening,
2. preclinical biomarker selection,
3. TCGA/TARGET Anderson-Darling concordance filtering,
4. clinical validation on CTRDB.

In the DROMA ecosystem:

• DROMA.Set provides the data structures and database access layer.
• DROMA.R provides the statistical analysis functions.
• DROMA.Meta is the application-layer workflow package that orchestrates those lower-level capabilities into an end-to-end biomarker discovery pipeline.

This makes DROMA.Meta an application package for DROMA.R, focused on running a standardized translational meta-workflow rather than exposing a broad new analysis API.

Position in the DROMA Stack

DROMA.Set  ->  DROMA.R  ->  DROMA.Meta
data layer     analysis     workflow application layer

• DROMA.Set: builds and manages DromaSet / MultiDromaSet objects.
• DROMA.R: performs drug-omics association analysis and batch screening.
• DROMA.Meta: turns those capabilities into a reusable workflow package for preclinical-to-clinical biomarker discovery.

What This Package Does

• Wraps the four-step DROMA meta workflow into reusable R functions.
• Keeps workflow logic in R/ as package-style function files.
• Lets users pass explicit runtime parameters such as:
  • drug
  • tumor_type
  • data_type
  • cores
  • db_path
  • tcga_rna_counts_dir
  • gene_probe_map_path
  • threshold settings for each workflow step
• Supports script-level looping over drug and tumor-type combinations through 01-run_meta_batch.R.

Core Entry Points

Package Function

The main workflow entry point is:

runMetaWorkflow(
  drug,
  tumor_type,
  feature2_type = "mRNA",
  data_type = "all",
  cores = 3,
  cell_min_intersected_cells = 20,
  pdcpdx_min_intersected_cells = 8,
  db_path,
  ctrdb_path,
  tcga_rna_counts_dir,
  gene_probe_map_path,
  output_base,
  override = FALSE
)

This function runs one full workflow for one drug and one tumor_type, and returns a one-row summary table while writing intermediate outputs to the workflow output directory. When override = FALSE, existing stage outputs under output_base/<drug>/<tumor_type>/ are reused and those stages are skipped.

Driver Script

01-run_meta_batch.R is the application driver script.

It is responsible for:

• loading package dependencies,
• loading local function files from R/,
• defining runtime parameters,
• reading valid_drugs.csv and valid_tumor_types.csv,
• looping over all selected combinations,
• writing a batch summary CSV.

This separation keeps the files under R/ closer to package-style reusable functions, while leaving actual experiment execution in a script.

Installation

Prerequisites

Install the lower-level DROMA packages first:

# devtools::install_github("mugpeng/DROMA_Set")
# devtools::install_github("mugpeng/DROMA_R")

Install DROMA.Meta

if (!requireNamespace("devtools", quietly = TRUE)) {
  install.packages("devtools")
}

devtools::install_github("mugpeng/DROMA_Meta")

Quick Start

1. Load the DROMA Stack

library(DROMA.Set)
library(DROMA.R)

2. Load DROMA.Meta Workflow Functions

source("R/FuncHelper.R")
source("R/FuncValidCheck.R")
source("R/FuncTcgaAD.R")
source("R/FuncMetaWorkflow.R")

3. Run One Drug and Tumor-Type Workflow

result <- runMetaWorkflow(
  drug = "Paclitaxel",
  tumor_type = "breast cancer",
  feature2_type = "mRNA",
  data_type = "all",
  cores = 3,
  cell_min_intersected_cells = 20,
  pdcpdx_min_intersected_cells = 8,
  db_path = "/Users/peng/Desktop/Project/DROMA/Data/droma.sqlite",
  ctrdb_path = "/Users/peng/Desktop/Project/DROMA/Data/ctrdb.sqlite",
  tcga_rna_counts_dir = "/path/to/tcga/rna_counts",
  gene_probe_map_path = "/path/to/gencode.human.v49.annotation.gene.probeMap",
  output_base = "workflow/Output",
  override = FALSE
)

print(result)

4. Batch Run by Script-Level Looping

Use 01-run_meta_batch.R as the application script.
Edit the runtime parameters in that file, then run:

Rscript 01-run_meta_batch.R

Repository Structure

R/
  FuncHelper.R
  FuncValidCheck.R
  FuncTcgaAD.R
  FuncMetaWorkflow.R
workflow/
  00-Eligible_Drug_Tumor.R
  01-Batch_Preclinical.R
  02-Select_Preclinical.R
  03-TCGA_AD_Filter.R
  04-Clinical_Validation.R
01-run_meta_batch.R
DESCRIPTION
NAMESPACE

Intended Use

Use DROMA.Meta when you already rely on DROMA.R for analysis and want a reproducible application package for a fixed biomarker discovery workflow.

Use DROMA.R directly when you need lower-level analysis functions, custom plots, or exploratory workflows.

Related Packages

• DROMA.Set: data structures and database layer
• DROMA.R: statistical analysis layer
• DROMA.Meta: workflow application layer