Skip to content

Results Page

This page documents the structure and organization of the Results Page analytical interface.

Purpose of the Results Page

The Results Page is the analytical workspace where processed data is displayed after successful upload and validation. It provides access to integrated database results and 56 analytical use cases organized into 8 specialized modules.

The page is dynamically generated based on uploaded KO annotations and remains accessible for the duration of the session (4 hours).

Global Organization of Results

Accordion-Based Layout

Results are organized as sequential accordion sections that expand on user interaction:

  1. Overview Card — Summary statistics (samples detected, total KOs, processing time)
  2. Database Integration Tables — Merged data display (BioRemPP, KEGG, HADEG, ToxCSM)
  3. Analytical Modules — 8 thematic sections containing 56 use cases total

Information Flow

Input (KO annotations)
  ↓
Validation (format, limits, encoding)
  ↓
Database Merging (BioRemPP + KEGG + HADEG + ToxCSM)
  ↓
Results Page Display (Overview + Databases + Modules)
  ↓
On-Demand Use Case Execution

Analytical Modules and Use Cases

The Results Page organizes analytical workflows into 8 thematic modules:

Module Focus Use Cases Primary Output Types
Module 1 Comparative Assessment (Databases, Samples, Regulatory) 6 Heatmaps, Stacked bar charts, UpSet plots
Module 2 Exploratory Analysis (Functional Potential Ranking) 5 Bar charts, Box-scatter plots
Module 3 System Structure (Clustering, Similarity, Co-occurrence) 7 Correlograms, Dendrograms, Scatter plots
Module 4 Functional and Genetic Profiling 13 Bar charts, Dot plots, Heatmaps, Radar charts, Sunbursts
Module 5 Modeling Interactions (Samples, Genes, Compounds) 6 Chord diagrams, Network graphs
Module 6 Hierarchical and Flow-based Functional Analysis 5 Sankey diagrams, Treemaps
Module 7 Toxicological Risk Assessment 7 Box-scatter plots, Chord diagrams, Density plots, Heatmaps, Treemaps
Module 8 Assembly of Functional Consortia 7 Frozenset visualizations, Heatmaps, UpSet plots

Total: 56 use cases across 8 modules

Module Structure

Each module contains:

  • Thematic focus addressing specific research questions
  • Multiple use cases (~8 per module)
  • Accordion-based layout (one use case per expandable panel)
  • Independent execution (no dependencies between modules)

Use Case Execution Model

On-Demand Execution

Use cases are not pre-computed. Each analysis:

  • Executes when user expands accordion and clicks "Run Analysis"
  • Generates visualization + data table dynamically
  • Returns results within the same accordion panel

Module Independence

Modules are conceptually independent:

  • No required execution order
  • Results from one module do not affect others
  • Users may analyze any subset of modules
  • All modules analyze the same merged dataset

Cross-Module Data Integrity

All modules use identical input data:

  • Same sample count across all analyses
  • Same KO coverage in all modules
  • Consistent database integration results

This ensures reproducible comparisons when interpreting results across modules.

Relationship Between Visualizations and Tables

Dual Output Structure

Every use case produces two complementary outputs:

  1. Interactive Visualization (Plotly-based)

    • Dynamic tooltips with detailed information
    • Zoom, pan, reset controls
    • Legend filtering (click to toggle series)

  2. Data Table (AG Grid or Dash DataTable)

    • Sortable columns
    • Searchable/filterable rows
    • Paginated display

Data Provenance and Consistency

Critical principle: Tables contain exactly the data used to generate the corresponding visualization.

  • No hidden filtering post-visualization
  • Sample traceability via "Sample" column
  • Parameter transparency (documented in use case description)

Display vs. Underlying Data

Display data (on-screen):

  • May apply Top N restrictions for clarity
  • Aggregated for visualization readability
  • Optimized for interactive exploration

Underlying data (available for export):

  • Complete dataset before visualization filters
  • All database fields retained
  • See Downloads section for export details

Session-Based Nature of Results

Data Retention Policy

Results are session-dependent and not persistent:

  • Storage mechanism: In-memory (Redis cache)
  • Lifetime: 4 hours of inactivity or until browser closure
  • Deletion: Automatic and irreversible upon session termination

User Implications

Session management:

  • No persistent storage of results
  • No retrieval from browser history
  • Keep browser tab open during active analysis
  • Re-upload required if session expires

Privacy and Security

Session-based architecture ensures:

  • Complete session isolation (unique session IDs)
  • No permanent user data retention
  • Automatic data purging after timeout
  • No cross-user data access

Limitations

BioRemPP results indicate genetic potential for exploratory analysis, not definitive biological activity or regulatory compliance.

For complete scope and limitations, see Limitations and Scope Boundaries.

Exports in Context

Results displayed on screen can be exported for external analysis and reproducibility. Export mechanisms include:

  • Database-level exports: Available in database section headers
  • Use-case-level exports: Available in use case headers

For complete documentation on export formats, scopes, metadata, and reproducibility requirements, see Downloads Guide.