I-SPI

A Shiny-based web application for analyzing and managing serology assay data with multi-user project management.

---

Overview

I-SPI is an interactive R Shiny application for processing, analyzing, and visualizing several serology assay types including bead-based immunoassay data, e.g. from ELISA, MSD and Luminex. It provides a unified platform for managing serology experiments with robust features for data import, quality control, curve fitting, and results visualization.

---

Features

User Management & Projects

• Secure Authentication: DEX OIDC integration via app.R and auth_config.R
• Multi-User Support: Role-based access (owner/collaborator) via user_management.R
• Project Workspaces: Isolated environments, switch & manage via ui_handler.R
• Access Control: Share projects with secure keys (UUID), validated in user_management.R

Data Import & Processing

• Multi-format Import: Handles various bead array file types (import_lumifile.R)
• Multi-Plate Support: Organize and analyze multi-plate experiments (combined_plates.R)
• Data Validation: Built-in QC checks (reset_import_values)

Analysis & Visualization

• Blank Controls: Background subtraction or sample gating
• Bead Count Metrics: Quality checks (bead_count_functions.R)
• Curve Fitting: Advanced algorithms (standard_curve_ui.R)
• Dilution Analysis: Comprehensive dilution series analysis (dilution_analysis_ui.R)
• Dilutional Linearity Correction: Interactive method for computing corrected interpolated values comparable between samples run at different dilutions. ('dilutional_linearity_ui.R')
• Outlier Detection: Statistical outlier analysis (outliers.R)
• Interactive UI: Dynamic tables, collapsible panels, real-time updates, export support

---

Technical Architecture

• Backend:
  • Database: PostgreSQL with schemas for users, results, and outliers (db_outliers.sql)
  • Robust DB Management: Transaction support, error handling, notifications
• Frontend:
  • Shiny: Main app in src/app.R
  • Responsive UI: Bootstrap, dynamic panels, interactive DT tables

---

File Structure

src/
├── app.R                 # Main app entry point
├── auth_config.R         # Dex OIDC config
├── user_management.R     # User & project operations
├── ui_handler.R          # UI rendering/event handling
├── import_lumifile.R     # Data import logic
├── outliers.R            # Outlier detection
├── curve_calc.R          # Curve fitting
├── *_ui.R                # UI modules
└── *_functions.R         # Analysis/processing functions

---

Database Schema

Defined in db_scripts/db_outliers.sql:

• i_spi_users: User, project, and permission management
• i_spi_results: Experimental data and results
• i_spi_outliers: Outlier analyses and context

---

Getting Started

Prerequisites

• R (≥ 4.0)
• PostgreSQL
• Required R packages (see source files)

Setup

1. Set environment variables in .env

Run the Application

# Install dependencies
# Configure database & Dex OIDC (Ability to disable it by setting flag = 1 Sys.setenv(LOCAL_DEV = "1"))
shiny::runApp("src/app.R")

---

Docker Support

• Dockerfile: Build application image
• docker-compose.yml: Orchestrate services

---

Security Issues

If you uncover a security vulnerability in this app or in its containerization, please do not create a public notice and description, but instead contact Michael.S.Zens@dartmouth.edu and he will return your email as soon as possible. If necessary our I-SPI team will provide a full team response to address this type of problem swiftly and rigorously.

---

Contributing

See the section: CONTRIBUTING

---

For questions, contributions, or issues, please open a GitHub issue or pull request!