🛣️ Roadmap

GeoChemistry Nexus aims to be more than just a "plotting software"—it is gradually becoming an integrated data · diagrams · calculation · collaboration toolchain for geochemistry and petrology research.

✅ Completed

Core Workflow

• Home: Quick links, practical widgets
• Data Preprocessing: Major oxide column recognition, Fe valence estimation / back-calculation, outlier, missing value, and detection limit handling strategies
• Discriminant Diagrams: Plot ternary diagrams, scatter plots, spider diagrams, etc. based on template library; supports data projection and style editing
• Geothermometer (GTM): Built-in multi-mineral calculation templates and Excel-like custom functions
• CIPW Norm Mineral Calculation: Whole-rock major element norm calculation and result export

Platform Capabilities

• Template Ecosystem: Official templates updated dynamically from the cloud; JSON / ZIP format custom template creation, import, and packaged sharing
• Multilingual: Interface localization (Chinese / English / German, etc.); one-click multilingual switching for diagram templates
• Export & Integration: PNG / JPG / BMP / SVG export; supports collaboration with third-party software such as CorelDRAW, Inkscape, and Adobe Illustrator

🔥 In Progress

• Continuous expansion of diagram template library
  Focus on adding geochemical discriminant diagrams such as tectonic setting discrimination diagrams and rock classification diagrams.
• Geothermometer module improvements
  Expand commonly used geothermometer / geobarometer templates.
• Documentation & localization
  Synchronously update Chinese / English / German and other documentation to reduce onboarding costs for new users.

📅 Near-Term Plans

Isotope Geochronology Plotting

• U-Pb concordia diagram (Concordia)
• Isochron diagrams for Rb-Sr, Sm-Nd, Lu-Hf, U-Pb, etc.
• Initial isotope ratio related diagrams (e.g., (⁸⁷Sr/⁸⁶Sr)ᵢ vs εNd(t))

🔭 Long-Term Vision

Machine Learning-Assisted Analysis

Introduce common machine learning workflows to assist geochemical data discrimination and classification, outputting model evaluation, variable importance, and visualizations such as ROC curves and confusion matrices—without requiring users to configure a Python environment themselves.

"New Diagram"-Style Discrimination Models

Support loading extensible machine learning discrimination models (e.g., new-type discrimination diagrams trained on big data), allowing researchers to directly apply "new diagrams" for analysis without mastering the full ML development stack.

AI Research Assistant (Phased)

1. Phase 1: RAG Q&A based on built-in help documentation and knowledge base, providing operation suggestions based on keywords and scenarios
2. Phase 2: Further assist data processing and analysis workflows, simplifying repetitive operations

Cross-Platform & Other

• Evaluate Avalonia and other options, gradually implement Linux / macOS support (currently Windows-focused)
• Continuously optimize performance, template formats, and developer extension interfaces