To an important degree, scientific research advances as a direct result of the tools scientists have at their disposal. Telescopes, microscopes, MRI machines and countless other devices have revolutionized one area of human inquiry after another. You don’t have to be a scientist to see that the game-changing tools of our time are computers and the software applications that run on them, but maintaining and building software is not the kind of thing that draws a lot of funding interest.
That’s why the Chan Zuckerberg Initiative started its Essential Open Source Software for Science program, so far awarding $8.8 million in grants in two funding cycles. Its goal is to support applications crucial to biomedical research, in particular.
Software may not sound quite as exciting as, say, a cure for cancer or any number of other chronic diseases that philanthropists have been eager to take on. Much of the motivation behind philanthropy for medical research is, in fact, highly personal and emotional, often in retaliation or defense against an ailment that has hit close to home. Software’s broad applications make it less likely to connect to a donor’s personal mission. More glamorous research tools, like high-powered microscopes and telescopes, have attracted their fair share of wealthy patrons.
But software is as essential to research as anything else in the lab. And in this era of global pandemic, when it is safe to say that more scientists than ever before are working on the same topics, the research applications that enable transparency, reproducibility and collaboration matter to everyone.
“These are absolutely critical pieces of infrastructure for science, yet they have never received dedicated funding support,” said Dario Taraborelli, CZI science program officer for open science. “That’s the crux of what we’re trying to address with this program.”
The first round of grants under the program, $5 million in total, was announced late last year. A second $3.8 million was awarded a few weeks ago, and a third round is on the way. According to CZI, the grants will help awardees hire developers, improve documentation, address usability, improve compatibility, onboard contributors and convene communities, among other functions.
Most of us are familiar with the names of common software applications we use at work and in our personal lives. But most of us are utterly unfamiliar with common applications used by scientists. Consider Matplotlib, an application that enables the creation of graphs, charts and other figures. Unless you’re in the research business, you’ve probably never heard of it, yet it’s one of the most widely used tools in science.
Taraborelli noted the lack of recognition that software tools like Matplotlib garner compared to research articles, for example. The more frequently an article is cited by other scientists, the more acknowledgment, career opportunity and funding the article’s authors receive. Not so for the creators and maintainers of open software applications. Matplotlib has been named as a “dependency” (the software equivalent of an article citation) more than 200,000 times; if it were a research article, it would be considered one of the most important publications in the history of science.
For the first time, CZI’s grant will give Matplotlib’s lead developer significant funding to work on and maintain the software. Other awards will support projects similarly important to research, with names similarly inscrutable to most laypeople, such as SciPy, QIIME 2, Bioconductor Build System, and ITK-SNAP.
CZI’s open software program is consistent with its broader approach to advance health and other areas by promoting the use of technology to make science more collaborative, transparent and reproducible—rather than by focusing solely on one specific problem or solution. For example, a couple of years ago, the funder partnered with UMass Amherst’s Center for Data Science to develop an artificial intelligence-powered program to analyze previously published research findings.
That tech focus makes sense, considering donor Mark Zuckerberg’s connection to digital tools, namely a little website called Facebook. Zuckerberg and Priscilla Chan founded CZI as a mission-oriented LLC in 2015, and it’s quickly become a prominent funder of science and education, and most recently, justice and opportunity. The outfit has also attracted some unwanted notoriety recently, facing demands for greater inclusivity and criticism from grantees over the way Zuckerberg is running Facebook.
While CZI has made tech tools a signature niche, some other big funders of science, including the Gates Foundation and Wellcome Trust, have also championed technology-based efforts to improve and empower research and collaboration within science. And the Simons Foundation established the Flatiron Institute to advance science through the development and improvement of computational tools.
Open source software applications may not sound as glamorous to lay philanthropists, but these applications are, as Taraborelli says, crucial throughout the endeavor of scientific research. And while some of the software projects that CZI supports under the program are specific to a particular field, others are foundational across many areas of study.
“As science becomes more and more computational, it has opened up entire datasets that were unthinkable just a few decades ago,” said Taraborelli, citing areas such as genetics that require big number crunching. The software enables more ways to analyze and learn from biomedical and other types of data, and allows more researchers to work on it. “But all of this depends on the ability of scientists having these tools, making sure these tools are well-supported and making sure people know how to use them.”
