Barcoding Results Dashboard

A Shiny-based interactive dashboard for visualisation, filtering, and downloading validated barcode sequences (and associated statistics) from the Barcode gene Extraction and Evaluation from Genome Skims & Evaluator (BeeGees) pipeline. Barcode validation is undertaken using the barcode_validator tool. Both BeeGees and barcode_validator were built for the Biodiversity Genomics Europe (BGE) consortium.

Quick Start

Prerequisites and Setup

1. Download R and RStudio
2. Install required packages:

install.packages(c("shiny", "DT", "jsonlite", "dplyr", "tidyr", "plotly", "shinyjs"))

Running the Dashboard

1. Download or copy launch_dash.R to your local directory.
2. Open launch_dash.R in R/RStudio.
3. Click 'Run App' (top right of the text editor). The dashboard will open in your default web browser. OR
4. Visit the deployed app.

Features

Search & loading options:

• Plate selection: Select and browse results for Process IDs in specific plates (BGE_XXXXX format).
• Project selection: Filter results by project codes to view all Process IDs containing specific project identifiers (e.g. BSNHM, BSNTN, BIOSC, etc).
• Progress tracking: Data loading progress and status updates.
• Data validation: Logging and error reporting for missing data.

Barcoding outcome overview:

• 🟢 Green - Pass: Both structural and taxonomic validation successful.
• 🟡 Amber - Partial: Either structural OR taxonomic validation successful (still considered overall fail).
• 🔴 Red - Fail: Neither structural nor taxonomic validation successful.
• barcode_validator validation criteria:
  • Structural: Filter to sequences with ambig_original == 0 (no ambiguous bases in original sequence (evidence of chimeric sequences)), stop_codons == 0 (no stop codons), reading_frame in [1,2,3] (valid reading frames), min(ambig_basecount) (fewest Ns in HMM-extracted barcode region (these Ns represent gaps in the barcode consensus sequence)), and with max(nuc_basecount) (longest sequence)
  • Taxonomic: Exact word match between family rank in BOLD (expected taxonomy) and taxonomic databases (observed taxonomy).
• BOLD systems integration: Clickable Process ID links to corresponding BOLD database records.
• Summary statistics: Pass rates and outcome distributions with visual summaries.

FASTA Sequences:

• Barcode sequences: View and download consensus barcode sequences in FASTA format.
• Export options: Download FASTA files or copy sequences to clipboard.

Interactive visualisation(s):

• Scatter plots: Interactive plotly charts with customisable X/Y axes across all data types.
• Bar charts: Process ID overview with proportional display options for fastp metrics.

Supported Data

The dashboard can filter results by plate ID, with 451 plates supported:

e.g., BGE_00001, BGE_00002, BGE_00003, etc.

Or by the following (27) project codes:

BSNHM, NHMCG, BGETR, BSUIO, BGLIB, BSNTN, BGEGR, DTAUT, HMAUT, 
BGENL, DTKNU, BBIOP, BHNHM, UNIFI, DTULO, MEAMP, MUSBA, BGSNL, 
BGSNH, BGEPL, EBGEP, BSCRO, BIOSC, INVBG, BCEMI, ILECA, ALPFU

Workflow usage

1. Select data: Choose either:
   • Plate: Select a specific plate from the dropdown (e.g., BGE_00001)
   • Project: Select a project code to view all Process IDs containing that identifier
2. Load results: Click "Load Results" to import and process the data.
3. View outcomes: Navigate to "Barcoding Outcome" to see:
   • Traffic light classification of all Process IDs
   • Summary statistics and pass rates
   • Clickable BOLD Systems links
4. Explore data: Use additional tabs to examine detailed metrics:
   • Barcodes: View and download FASTA sequences with filtering options
   • BeeGees summary statistics: Pipeline summary results with interactive plots
   • Barcode validation: Detailed structural and taxonomic validation data
5. Export results: Download filtered data using the export buttons on each tab.

Support

For issues or questions, please open an issue on this repository.