Attribution requirement

If you use data downloaded from PHAROS, you must properly credit its source. This is the only formal requirement for using our database at this time.

Required: cite downloads

Every download is assigned a unique permanent link, which shows the list of associated contributors and projects.

This link must appear either in the full text or references of the paper.

We prefer that you include the web link in the methods section, data availability statement, or acknowledgements (i.e., indexable text; not supporting information) of all publications using that download. For example:

... To understand coronavirus dynamics in deforested and intact landscapes, we downloaded all records of bat coronavirus surveillance from the Pathogen Harmonized Observatory (PHAROS), searching by country for "Brazil" and by taxon for "Coronaviridae" and "Chiroptera"; our search returned a total of 86,753 results. (Data are permanently available on the PHAROS site:

Data availability statement: All surveillance data used in this study are available from the Pathogen Harmonized Surveillance database (

Depending on journal style, you can also cite data downloads in your methods; for example:

PHAROS (2023). Accessed on 2023-08-11 21:28:48 UTC.

Suggested: cite projects

If your work focuses on re-analysis of a specific project or handful of projects, we suggest also individually citing those projects, in addition to citing downloads. Again, these can be cited in-text or in the references:

In this study we reanalyze data on paleolithic nematodes collected by Pumpkins et al. and openly shared on the Pathogen Harmonized Observatory web platform for wildlife disease (

David Pumpkins, Robert Moynihan, Michael Day. Parasite detection using ancient DNA from skeletons (2023). Accessed on 2023-10-24 17:32:09 UTC.

Frequently asked questions

Q: What happens to cited datasets?

For now, citing your data sources is just a best practice for findability and attribution. Eventually, a top priority for our developers is to add a webscraper to our back-end that tracks how projects are being cited and automatically populates that information onto researchers' project pages, so that they can show off how their data are being used.

Q: Why don't we use DOIs?

This might not be true forever, but in our beta launch:

  • Working with DOIs requires a fair bit of longevity planning (which we will do anyway)

  • Experience with other major open data sources tells us that DOI citations are often skipped by end users, partly due to the challenges of data citations and journal reference formatting. (This is why we want you to have the option to put the token in-text - and why there's no excuse not to "cite" your data that way.)