Repository Cleanup & bayesTPC Updates (June 2025)

June 12–14, 2025: bayesTPC Function Updates • Updated to the development version of bayesTPC from GitHub (johnwilliamsmithjr/bayesTPC). • Renamed data loading function: • get_datasets() → get_VB_datasets() • Removed deprecated function bayesTPC_summary(); replaced with standard S3 method: summary(). • Posterior prediction functions (plot_prediction, posterior_predictive) now return custom objects (e.g., "btpc_prediction"), requiring manual extraction (e.g., $summary). • plot_prediction() returns NULL unless temp_range is explicitly provided.

⸻

June 29, 2025: Model Fitting Error & NIMBLE Compatibility • Encountered error with b_TPC() due to missing internal getNimbleOption() function in newer versions of nimble. • Fixed by downgrading nimble to version 0.13.2:

remotes::install_version("nimble", version = "0.13.2")

⸻

June 30, 2025: GitHub Push Failed Due to Large Files • Updated .gitignore to exclude cache, site output, and large intermediate files:

*_cache/ _site/ *.rdb *.rdx *.RData *.tar.gz

•	Removed large files from Git history using git filter-repo:

git filter-repo --force --path-glob '*.rdb' --invert-paths

•	Reconnected origin remote and force-pushed cleaned history:

git remote add origin https://github.com/VectorByteOrg/bayesTPC-tutorials.git git push --set-upstream origin yusi --force

⸻

August 19, 2025: Site Reorganization • Merged Getting Started and Materials pages to eliminate redundant content • Updated navbar: "Materials" → "Tutorials" • Added subsection anchors to EEID tutorials for direct linking • Integrated Quick Start guide with working code example • Added tutorial descriptions with time estimates and learning paths

⸻

December 2025: Complete BTV Refitting Implementation • Successfully implemented comprehensive BTV refitting pipeline for El Moustaid et al. (2021) • Built prior translation system to bridge paper notation and bayesTPC expectations • Created robust data processing pipeline for Appendix A.6 data • Implemented MCMC fitting with custom priors from Appendix A.2 • Generated thermal performance curves for 8 trait-species combinations • Reproduced Supplement Fig. A.2 uncertainty attribution analysis • Created comprehensive visualization gallery with HTML interface • Integrated results into main Bluetongue Refit page with blog-style narrative • Added quality control and convergence diagnostics • Documented technical challenges and solutions

Key Results: • Successfully fitted p, b, ν, ρ traits for C. sonorensis and C. variipennis • Development rate (ν) contributes most uncertainty to derived quantities • Vector competence (b) contributes least uncertainty • Optimal temperature ranges: 7-30°C for survival, 35-40°C for competence • All fits achieved R-hat < 1.1 and ESS > 1000

Files Added: • Complete R pipeline (priors_translate.R, fit_trait.R, etc.) • Data files and configuration (priors_tableA2.yaml) • Output files (fits, plots, summaries) • Visualization gallery (HTML interface) • Updated refit_guide.qmd with comprehensive results