Where did Rxivist go?

March 1, 2023

Rxivist.org was an application that combined metadata for all preprints posted to bioRxiv and medRxiv with information about Twitter.com to showcase the most discussed preprints in biology. A lot has changed since we started the project in 2018—the biggest difference is that bioRxiv now has a straightforward API that enables programmatic access to much of the metadata previously only available from Rxivist. The other critical change is that Crossref, our source of social media data, lost access to all information from Twitter in February 2023. Retrieving data on hundreds of thousands of preprints directly from Twitter would cost thousands of dollars per month according to their old pricing, and their product plan is now too volatile to depend on specific features remaining available. These factors, combined with a lack of funding for the project, have led us to conclude that ongoing maintenance of the Rxivist website and crawlers is no longer practical.

We're very grateful to Crossref, the bioRxiv team (who, we should note, were very supportive but had no affiliation with Rxivist), and all the people we got to talk to and work with over the course of this 4.5-year effort. Information from the Rxivist project is still available in several ways:

• The final database snapshot is available on Zenodo, along with all previous versions of the database. (This contains all data used in the Rxivist web application.)
• "Tracking the popularity and outcomes of all bioRxiv preprints," our 2019 paper about the makeup of bioRxiv.
• "Rxivist.org: Sorting biology preprints using social media and readership metrics," our 2019 paper about the Rxivist web application.
• "International authorship and collaboration across bioRxiv preprints," our 2020 paper about worldwide participation in the preprint ecosystem.
• The rxivist GitHub respository contains the code used for the API, the programmatic interface that was also the backend for the Rxivist website.
• The rxivist_web repository contains the code used for the Rxivist website.
• The rxivist_spider_biorxiv repository contains the code for the web crawler used to index preprint metadata from bioRxiv and Twitter.
• The Panlingua repository contains the code used for the Panlingua website, which interacted with Google Translate to enable searching and reading of bioRxiv preprints in dozens of languages.

That's all for now. Thanks for all your support.

—Rich Abdill
Blekhman Lab, University of Chicago