Computational biologist Geoffrey Siwo is out to ensure that the groundbreaking gene-editing tool will be available to scientists — and eventually patients — everywhere in the world.
From cancer to malaria to HIV, CRISPR is set to open up all kinds of dramatic breakthroughs in medicine, as well as agriculture. But who gets to wield this exciting new gene-editing tool, and who will benefit from the outcomes? Systems biologist and TED Fellow Geoffrey Siwo — who works at the Eck Institute of Global Health at the University of Notre Dame — wants to make sure that this emerging technology gets an equitable start.
Last week, Siwo published new research that he hopes will address and prevent gender, geographical and other disparities in CRISPR research. His paper, The Global State of Genome Editing, used algorithms to text-mine recently published scientific papers on genome editing, analyzing the global and gender distribution of nearly 10,000 researchers in the field as well as identifying the genes and animal or plant species these papers focused on.
Here, Siwo shares some of the disparities that his new paper highlights and tells us why it’s so important to start making genome-editing research more accessible to all.
What’s in this new paper, and why is it useful?
This is the first paper that takes a global view of about 2,000 papers published by nearly 10,000 researchers across the globe on two genome-editing technologies — covering a period from the beginning of 2017 to May 2018. It would take a long time for a single person to read 2,000 papers, let alone comprehend what’s going on across all of them. So I turned to computational algorithms to mine the texts and automatically extract key information, such as the names of the researchers, their institutional affiliations and the genes, diseases and species being targeted by genome-editing technologies in research. We also used computational approaches to infer such information as the gender of researchers and the locations of the institutions involved.
What kinds of disparities does it flag up?
All kinds of disparity are evident in our results. The first are regional disparities: institutions in the US and China are leading in terms of publications on genome-editing technologies, followed by Japan and the United Kingdom. Even within the US, you can see that there are more papers emerging from the West and East Coasts, and fewer publications coming from other parts of the country. Africa-based scientists are heavily underrepresented, even though malaria is one of the top diseases for which genome-editing technologies are being explored as a means to eradicate the disease.
The paper also reveals gender disparity. On a positive note, when you look at first authors (i.e., the people who perform the technical aspects of the research) in genome-editing publications, you’ll find nearly equal representation between male and female authors. Sadly, when you look at the last authors — who are normally the principal investigators — you’ll find that only about 26% are females, while the remaining 74% are males. (These figures are based on an algorithm that predicts gender of authors in papers based on their first names, so they’re estimates.)
“If scientists in Africa can read open-access papers about CRISPR but have no means to use it in experiments, they still can’t transform that knowledge into technologies that will improve the health of Africans.”
Principal investigators determine the direction of research being done in a lab, so we have to ask ourselves what it means when those leading genome-editing research are mostly male. For example, gender diversity has been shown to be essential in enhancing scientific discovery. Furthermore, addressing women’s health using genome-editing technologies would be futile without participation from women scientists, as they’re more likely to recommend research on women’s health.
Our algorithms also detect which papers are published in open-access journals. This is important because this determines who reads the papers. Some published papers are not open access — problematic because if only those who can read these are those who can afford them, it affects who ends up developing the technology. Resources play a part, too: if scientists in Africa can read open-access papers about CRISPR but have no means to use the information in experiments, they still can’t transform that knowledge into technologies that will improve the health of Africans.
Why is it especially important for scientists from Africa and other parts of the developing world to pay attention to CRISPR right now?
CRISPR has the potential to have a radical impact on many diseases, including those which could not previously be treated—especially genetic diseases. But it’s also a huge opportunity to solve the problem of health disparities between rich nations and poor. If, as it emerges, the technology itself is developed in a biased way, then different populations will not benefit equally, whether economically or medically.
“CRISPR is also a huge opportunity to solve the problem of health disparities between rich nations and poor.”
For instance, there is evidence that CRISPR works differently on different human genomes. So if only scientists from a certain part of the world — such as Europe and North America — are involved in using a technology, then those scientists are naturally likely to be biased in their choice of which diseases to study and focus their efforts on local populations, and people in other parts of the world will lose out. Meanwhile, national funding agencies may have to base their budget allocations on local priorities. This is research disparity.
What scientific proof is there suggesting that there could be differences in how CRISPR technology works on different human genomes?
On the one hand, CRISPR is so powerful because it can be very precise. But that precision could mean unexpected consequences when the technique is applied to people you’ve never sequenced before. So it’s a double-edged sword.
Feng Zhang, who is credited alongside Jennifer Doudna with driving CRISPR as a genome-editing technology, has done a study showing that genetic variation in human genome sequences can have an impact on CRISPR’s editing outcome. In other words, if the CRISPR sequence was designed with one person’s genome in mind, and if you apply the same edit to another person’s genome, a side effect could result. More recently, another group from Harvard University scanned a set of more than 7,000 human genomes from different world populations and reported similar observations.
“If the CRISPR sequence was designed with one person’s genome in mind, and if you apply the same edit to another person’s genome, a side effect could result.”
The results of the two studies suggest that it’s advisable to sequence the genome of the person being treated before applying CRISPR technology to them. That’s one way of addressing the problem. From my point of view, another way to address the problem is to gather as much genetic data as possible, from as diverse a population as possible, and then build algorithms that learn from this data to develop better CRISPR designs. Developing these kinds of algorithms is part of my ongoing research.