Open science to tackle plant health emergencies: enough excuses, please!
The thorny politics of the global food trade are often an excuse against the application of open science principles to plant disease epidemics. But scientists have often themselves to blame for the slow release of information and data, and there are ways around some of the issues.
Cite as: Kamoun, S. (2022). Open science to tackle plant health emergencies: enough excuses please. Zenodo https://doi.org/10.5281/zenodo.6857665
Imagine there is an outbreak of some nasty disease in your neighbourhood. Say for instance the rare and frightening Ebola virus disease. And imagine that you only hear about this after a couple of years, even though your local doctors and hospitals were very much aware of the outbreak. How would you feel about that?
This situation happens in plant pathology. Over and over. Here is two recent examples.
Sicily stem rust outbreak
Stem rust is a major disease of wheat that can cause major crop losses. Until recently, the disease has more or less vanished from Europe, but sporadic signs of the wheat stem rust were reported in Germany and other northen European countries from 2013. This was worrisome enough, but the climatic conditions in northern Europe aren’t generally viewed as conducive to the heat loving pathogen and destructive stem rust outbreaks may remain limited, at least in the near future. What truly concerned plant pathologists was for the pathogen to reach the warmer climes of Southern Europe and North Africa. Following the early reports of 2013, everyone was on the alert.
Common sense would, therefore, call for a massive outbreak in Southern Italy to be instantly reported with an alert issued to farmers and agronomists without any delays. None of that happened when Dr Biagio Randazzo detected stem rust on April 19th 2016 in experimental wheat plots at Ciminna in the Palermo province of Sicily. The outbreak ended up damaging thousands of hectares of durum and bread wheat in Sicily. But radio silence until early 2017.
The sicily stem rust outbreak was only made public on February 2, 2017, 10 months after Dr. Randazzo walked through the infected wheat fields. The alert document titled “CAUTION: Risk of wheat stem rust in Mediterranean Basin in the forthcoming 2017 crop season following outbreaks on Sicily in 2016” was accompanied by spectacular photos Dr. Randazzo took in May 2016. They carried the dramatic caption: “Heavily infected fields of commercial durum wheat, Sicily.” The warning read as follows:
“The epidemics were estimated to cover several thousands of hectares resulting in high inoculum load that could pose a threat to surrounding wheat areas in the forthcoming 2017 crop season, if environmental conditions prove suitable. Growers in at risk areas should be aware of the possible risk (on both durum and bread wheat), monitor crops for the early appearance of stem rust and undertake timely control if necessary.”
Sicily is just hundreds of kilometers away from Tunisia, my North African homeland, meaning that I have personal reasons to be pissed. Wheat is the principal staple food in North Africa, adding up to about half of all consumed calories. North Africans cannot think of a meal without a loaf of bread, whether it’s the traditional flattened type or the adopted French baguette. Evidently, the wheat crop is of huge strategic relevance to food security and political stability in the region. North African agronomists and farmers deserved better than wait a full 10 months before hearing about the outbreak. But that’s exactly what happened.
I’m not privy to the reasons behind these delays. My understanding is that it was deemed necessary to determine the race of the stem rust pathogen — an indication of how dangerous the strain can be — before raising the alert. Did this justify the delay? Wasn’t it already obvious that this was a massively destructive outbreak, different from anything seen in the region in recent history?
Wheat blast in Zambia
Wheat suffers from another awful disease, wheat blast. And this disease has been on a rampage, having spread out in 2016 from South America, where it first emerged in the 1980s, to South East Asia, specifically Bangladesh. Plant pathologists have been on the alert for the presence of the pathogen in other countries. Because the wheat blast fungus can be transmitted through seeds, the risk of spread through the wheat trade is real. In a 2018 technical article in Annual Reviews of Phytopathology, Paulo Ceresini and co-authors went through the trouble of listing the 65 countries that import wheat from Brazil but have not reported wheat blast. The map reads like a doomsday scenario for the spread of this pathogen. Africa was heavily represented among the 65 at-risk countries.
Therefore, you can understand how I almost fell off my chair when I saw a paper in September 2020 that describes wheat blast in Zambia from outbreaks that happened in 2017–2018, a full 2 years after the disease was detected. The article published in the journal PLOS ONE was submitted on March 22, 2020 and accepted by the journal on August 21, 2020. Yet somehow, it wasn’t deemed necessary to raise the alert until after its formal publication on September 21, 2020. For one thing, the article could have been posted in bioRxiv, the scientific depository that aims to accelerate the dissemination of scientific information. Preprinting the article in March 2020 would have in no way delayed formal publication in the journal given that PLOS, the publisher of PLOS ONE, encourages posting preprints just like the majority of scientific publishers. If you don’t believe me, check this list of academic publishers by preprint policy. It’s all Unrestricted / Green these days, and we can all be proud of having collectively pressured publishers to adopt preprinting as standard procedure.
Preprinting is the least we can do when dealing with emergencies. This platform for rapid sharing of biological knowledge has seen exponential growth ever since bioRxiv went live in 2013. It truly came of age during the COVID-19 pandemic. By facilitating the exchange of scientific information on the COVID-19 pandemic, “preprints have changed science forever”, wrote Nature Medicine. Preprints essentially shaved months from the time it traditionally took for a scientific article to reach its audience. When flawed preprints were posted, the community immediately reacted and pointed to the issues. It was open and live peer-review. It’s how science should operate in the digital era.
An earlier alert would have saved months for the scientific community to come together and help generate useful information about the African wheat blast outbreak. This is exactly what we did when we finally heard about the outbreak in October 2020. A volunteer collaborative effort that ended up including 10 different institutions, including our colleagues in Zambia and Bangladesh, determined within weeks that the African outbreak was caused by the same aggressive pathogen strain that has been wreaking havoc in Bangladesh. Further analyses showed that the Asian and African pathogen populations were probably the result of two independent introductions from South America. The results of these analyses, which were first published in the scientific data repository Zenodo under the umbrella of the OpenWheatBlast initiative, were more recently combined into a full bioRxiv preprint. The paper lists the full set of datasets and preliminary reports as shown below, and all authors were invited to join in the full-length community paper.
Why are some plant pathologists reluctant to embrace open science?
Two main arguments pop up whenever the topic of open science is presented to the plant pathology community.
First, governments are reluctant to be open about new outbreaks for fear that it would stall their capacity to export the affected crop.
Second, there are no incentives for academics to embrace open science in a world dominated by superficial, some would say idiotic, research assessment metrics like journal impact factors.
Although both points have a certain degree of validity, I think their importance is exaggerated and neglects recent trends both in the global response to infectious diseases — as illustrated by the COVID-19 pandemic — and progressive developments in scientific publishing and research assessment. Yes, the complicated politics of the global food trade does interfere with the free exchange of information, scientific or other, and nobody is being asked to break the law of the land. But there are too many examples where the scientists seem to be part of the problem and the governments become a convenient excuse for a tendency to hoard information and data, and to only share them with the community after publication in a journal.
An action list to promote the application of open science to plant pathology
To move things forward and address the prevailing issues, I list below a list of actions that any of us could undertake. As always, I strongly believe in the adage: evil prevails when good people do nothing. Let’s all try as much as we can to do some of these things to build further momentum for the embrace of open science in plant pathology.
1. A more systematic and firm commitment to open science among scientists and scientific societies; promote the International Society of Plant Pathology “Code of Ethics”; lobby your national plant pathology society to embrace the code of ethics and similar initiatives.
2. Incentivise open science and alternative forms of publishing in your academic spheres; ensure that your institution signs up to the San Francisco Declaration on Research Assessment (DORA); call out DORA signatories whenever they don’t respect the chart, e.g. by promoting Journal Impact Factors and basing decisions on where the papers were published.
3. Contribute to training and education about open science, preprinting, and new forms of publishing; educate your students and colleagues about these concepts; point them to websites like ASAPbio and the Wikipedia List of Academic Publishers by Preprint Policy.
4. Educate governments about the importance of sharing data and the negative consequences of withholding crirical information about plant health emergencies; work with your local stakeholders to explain the importance of rapid responses in the case of plant health emergencies; use the COVID-19 pandemic as an example of what can go right and what can go wrong.
5. Increase plant disease surveillance at points of entry given that importing countries, unlike exporters, are incentivised to detect new pathogen incursions; use COVID-19 as an example of the impact of transparency, or lack of, on a country’s reputation; most countries don’t want to become a pariah state.
6. Promote a more open discussion about the impact of plant germplasm movement in the introduction of new diseases; move beyond treating this topic as taboo.
7. And finally, lead by example; don’t hoard your data and share it whenever it’s realistic especially if you work on epidemic diseases; educate your students and collaborators about the fact that publishing your data as a doi preprint or dataset will give you full credit and will not prevent you from publishing later on in a journal.
A code of ethics for plant health emergencies.
Check the OpenPlantPathology Community.
Get your institutions to sign up to The San Francisco Declaration on Research Assessment @DORAssessment
Preprints do not prevent you from publishing in journals. Check this List of academic publishers by preprint policy. It’s all GREEN these days!
Kamoun, S., Talbot, N.J., and Islam, M.T. 2019. Plant health emergencies demand open science: Tackling a cereal killer on the run. PLOS Biology, 17:e3000302.