[jaempawi]

Summary

New look for the FunGen-xQTL Resources site, built as layouts over the hugo-book theme. Replaces #52, which merged an empty commit. Authors still edit the same Markdown files.

Site

• Home page with the flagship paper's numbers (77 molecular datasets, 8 QTL modalities, 2,328 participants) and "Browse by" cohort, QTL modality and AD GWAS.
• Data catalog with a cohort x modality grid, filters and a dataset table (cards on phones).
• Dataset pages open with a summary card: cohort, data type, tissue or cell type, sample size, status, access, lead analysts and Synapse link.
• New pages: How to get the data, Dataset inventory, Analysis coverage.
• Dark mode and print styles; "Propose a dataset" issue form.

Naming

• glycoQTL instead of gpQTL in page text, links and the menu. Synapse file names and analysis-repo paths that contain gpQTL are unchanged.
• snRNA-seq datasets named CUIMC1 and MIT instead of by lab; PI names unchanged.
• snATAC multiome data named CUIMC2, as in the flagship paper.
• Modality labels follow the paper (glycoprotein abundance, metabolite abundance, histone acetylation).

Data access

• Controlled Synapse data: request data access on the AD Knowledge Portal.
• NIAGADS: NG00184 xQTL summary statistics and fine-mapping are open access with no application. Controlled individual-level data are requested through NIAGADS DSS by a PI with their Signing Official, with eRA Commons IDs and Login.gov or ID.me identity verification.

Metadata headers

• Dataset pages get a metadata header (type, modality, cohort, context, access, status, Synapse IDs, lead analysts) that drives the catalog and summary cards.
• Sample size, status, release and access are left blank where not yet confirmed; curators fill them in. Pages that already had a header keep their values.
• synapse_ids list only IDs the page itself cites.

Build

• scripts/hugo_generator.py copies site/ into the build and passes the headers to Hugo.
• CI installs PyYAML and pins hugo-book to 4074906.