Maintaining the Kissler Lab site

A practical guide to keeping this site up to date — written for Stephen and for future Claude sessions. The site is a custom Jekyll theme (no external theme); content lives in Markdown and _data/*.yml, and the design lives in _layouts, _includes, and assets/css/main.scss.

Run it locally

bundle install          # first time only
bundle exec jekyll serve # then open http://localhost:4000

Edits to .md, _data, _includes, and _layouts hot-reload. Editing _config.yml requires restarting the server.

Deploying

The live site builds from the master branch on GitHub Pages. This redesign lives on the redesign branch — merge it to master (or push to master) to publish. There’s no build step to run; GitHub Pages builds Jekyll for you.

Common edits

Add or update a publication → _data/publications.yml

Each entry looks like:

  - title: "Paper title"
    authors: "First Author, …, Stephen M. Kissler"   # order matters
    venue: "Journal or medRxiv/arXiv"
    year: 2026
    type: paper            # or: preprint
    tags: [viral-kinetics, surveillance]   # options are the `filters:` at the top
    senior: true           # true if Stephen is FIRST, LAST, or co-senior author
    url: "https://doi.org/…"

Keep the list newest first. Notes:

• senior is the important flag: the home page’s “Recent work” strip shows the six most recent senior: true papers. Set it true only when Stephen is first, last, or co-senior (e.g. the measles paper, where he’s second-to-last but co-senior). Middle-author papers get senior: false (or omit it).
• tags drive the topic filter chips on /research/. To add a new topic, add it under filters: at the top of the file, then tag papers with it.
• authors — verify against the DOI. Author lists are easy to get wrong from memory; the reliable source is Crossref: curl "https://api.crossref.org/works/<DOI>". Use … to elide a long list (e.g. "First Author, …, Stephen M. Kissler").
• Lab members’ names are bolded automatically — see below.

Bold a new lab member in author lists → _data/lab_names.yml

Any name in _data/lab_names.yml is wrapped in bold wherever it appears in a publication’s author string (via _includes/authors.html). When someone joins, add every form their name might take, e.g.:

- "Jane Q. Public"
- "Jane Public"

Add or update a person → _data/people.yml (+ a bio page)

• _data/people.yml has current: and alumni: lists (name, role, dept, photo, and a page: link). This drives the People index and the home “People” strip.
• Full bios are pages in _pages/people-<Name>.md using layout: person. Copy an existing one as a template. Put photos in assets/images/.

Update the news feed → _pages/home.md

The body of _pages/home.md (below the front matter) is the “Latest from the lab” list — plain Markdown, grouped by ### YEAR. Add new items at the top.

Edit navigation → _data/navigation.yml. Research themes → _data/research.yml.

Automated publication sync (ORCID → pull request)

.github/workflows/sync-publications.yml runs weekly (and on demand). It:

1. reads new DOIs from your ORCID record,
2. skips any already in _data/publications.yml,
3. pulls author list / venue / year from Crossref, and
4. opens a pull request with draft entries — tags: [] and senior: false left as TODO for you to fill in.

Nothing changes on the live site until you review and merge that PR. This keeps the automation reliable (ORCID/Crossref have real APIs — Google Scholar does not) while leaving the editorial calls (tags, senior authorship) to you.

One-time setup: add your ORCID iD as a repo variable — Settings → Secrets and variables → Actions → Variables → New variable, named ORCID_ID, value like 0000-0002-1825-0097.

Run it manually: Actions tab → “Sync publications from ORCID” → Run workflow.

Run it locally (to preview):

ORCID_ID=0000-0002-XXXX-XXXX python scripts/sync_publications.py
git diff _data/publications.yml

Where things live