1. Need

Humanity’s urgent need for a vaccine, or set of vaccines, for SARS-CoV-2 needs no elaboration. Quite simply, this is one of the more pressing scientific/humanitarian needs of the century.

2. Landscape

There is an enormous amount of ongoing work in the public, private, and nonprofit sectors. The bulk of it, however, is along business-as-usual lines, with heavy competition among companies, even nations. “A global arms race for a coronavirus vaccine is underway” reports the New York Times (“Search for Coronavirus Vaccine Becomes a Global Competition,” NY Times, March 19, 2020).

The standard approaches are characterized by research conducted in single labs; duplicative work; adherence to monopolistic intellectual property regimes, which tend to limit availability; concerns over affordability; private use of public funds; work driven by financial incentives, which often do not map to public health needs.

As death tolls mount, the bankruptcy of the traditional models is becoming apparent, and openness is coming to the fore. The urgency of the situation has forced changes. Players are collaborating more, and companies, often under pressure, are starting to release their hold on intellectual property and their claims to incentives, exclusivity, and economic rents.

The genome of SARS-CoV-2 was openly posted by scientists in China, although reportedly at great personal cost. This keeps with common practice of pathogen genomes being openly published.

A few respected groupings have been pursuing a more expansive approach for some years, doing terrific work. GLOPID-R, an international network of 28 or the world's major research funding organizations, insists that their grantees promptly share data on epidemic diseases. Established after the Ebola crisis, the global funding network CEPI insists on “equitable access.” CEPI was based on the insight that vaccine development for epidemic diseases was an area of market failure. CEPI co-founder Bill Gates observed that "[t]he market is not going to solve this problem because epidemics do not come along very often — and when they do you are not allowed to charge some huge premium price for the tools involved.” CEPI has many private sector industry partnerships; it funds companies but does not have rights to use the intellectual property developed by those companies, nor does it have full control over the availability of the ensuing products—note the controversies over German vaccine development company CureVax, a CEPI funding recipient, which was the subject of an imbroglio with US President Trump, who allegedly wished to buy the company and acquire its IP for US use.

3. The Gap: Private Open vs. True Open

If we had just arrived on this planet Earth, and were to design a system for vaccine R&D from scratch, optimizing only for scientific progress, what would it look like?

Science is based on sharing. Scientific advances rest on the shoulders of giants.

In a system that was serious about developing a vaccine as rapidly as possible, we would work together, across labs and borders. The minds of the globe would train themselves collectively on one pathogen, the SARS-Cov-2, share all the data, and work openly and rapidly and collaboratively, in public space, sharing ideas and discussions and analyses and hypotheses globally and rapidly, with results available for anyone to take forward.

In today’s crisis, with development of an accessible/publicly available vaccine being an overweening goal for humankind, would it not behoove us to try such an approach?

In Reinventing Discovery: The Age of Networked Science, from the Princeton University Press, Michael Nielsen describes open science efforts, and the rapid progress that can be made.

In some quarters, such as those backed by CEPI and GLOPID-R, there are elements of openness. We might call these efforts “private open.” They utilize open data sets (as well as private data sets), and share their results once finished.

But a much more open, and thus arguably faster and robust approach is possible, and, startlingly, has seemingly not yet been pursued, even though it coheres better with the way science should intrinsically proceed.

In open source software, not only are the incoming and outgoing results widely usable – bookends so to speak, but the very process of development itself, the actual daily work, is open, collaborative, global, and characterized by rapid exchange. It is the magic of this collaboration that has accounted for the creation, for example, of Linux, which dominates the supercomputer market, and together with its compatriots Apache, MySQL, and Perl, birthed the World Wide Web. The rapidfire give and take, and constant improvement, the multiplicity of ideas and hypotheses that can be proffered, tested and shot down by the global brain—all of this enabled open source software to be more rapidly developed, more robust, and in some instances flat out better than software made even in the largest of corporations.

Can we learn from these principles, and apply them to the process of vaccine R&D?

Of course vaccines are not software, and require wet labs, and clinical trials. And so the application is not lockstep. But the principles and techniques of open collaboration, open IP, heavy use of computation, crowdsourcing, massive open research and development, rapidfire commentary and exchange, could be utilized.

Imbuing not just the inputs and the outputs, but the process itself, to the maximum extent possible, with a global, open, crowdsourced collaborative approach, seems to have not been aggressively tried, and to bear promise, as it is consistent with the process of scientific discovery itself. Even if there were just a chance that it could yield a better, different result, would that not be worth trying, for a global crisis such as this?

Today, with much of the planet at home, and many working digitally, we are at a unique moment in time to explore such approaches.

1. Overview

The program would focus on in silico reverse vaccinology approaches, examining the genome of the virus itself, and of related viruses, to generate possible vaccine candidates.

This examination would take place via open, global, crowdsourced computational analysis. Results would be posted online for anyone in the world to take forward. With the current crisis, the takers would be many. Governments are searching for candidates to take into publicly-funded clinical trials.

Additionally, there is an iterative aspect. Once wet lab work is performed and insights obtained, there will be iteration, with further computational work to perform.

2. Phase Zero: Foundational Open Resources

1. Data

   • Immediate (1 week or less)

     • Assemble a web site that includes links to all relevant open data.

   • Midterm (1 week to 2 months)

     • Ultimately, this data should be searchable and interoperable.

   • Long term (2-3 months)

     • Knowledge base/knowledge graph, auto-updated by AI, of all relevant publications, with insights obtainable by a graphic visualization. OSPF/Mayo Clinic have together created such a tool for tuberculosis.

     • Include -omics, clinical, and public health data.

2. Computational Tools. Ultimately include on this web site, for those who lack them. In the drug context, OSPF has seen pro bono use of commercial discovery tools made available. In the vaccine context, there are several open source reverse vaccinology software tools available.

3. Community. The community of solvers here would of course include virologists and vaccine experts, potentially from academia, nonprofits, government, and industry, but could also include computer experts and potentially even citizens (e.g Fold-It, protein folding game). OSP’s existing community could be tapped. With the world waiting with bated breath for a vaccine, and a desire to help, in what is a sort of Apollo Project for our time, a carefully structured open call could potentially yield legions of people.

3. Phase One: Million Hypotheses

The world is trying to do years of research in a few months. There are many hypotheses to generate and pursue. It could be the out of-the-box, creative idea that hits the mark, as is often the case in science.

Here we would simply ask people to propose their hypotheses, which would be displayed publicly, with the opportunity for others to comment on them. Ultimately there could be a ranking system, with ID’s for those who are professional virologists and vaccinologists.

These hypotheses, and their associated commentaries would be displayed to the world to take forward, and could become essential reading for the vaccination R&D laboratories of the world.

We would use copyleft principles – so the ideas would be free to use, although attribution/credit of a nonmonetary type should be given to their creators.

4. Phase Two: In Silico Explorations

The community could then self-organize or self-initiate and conduct in silico explorations of these hypotheses. Guidance could be provided as well by professional researchers. The community would be also available to these researchers as needed. Some in silico explorations would be done iteratively, to refine hypotheses after wet lab work had been performed. Initially, such explorations would be done by community members themselves using their own resources. Later, this effort could provide open source and possibly commercial software tools to the community.

5. Phase Three: The Beyond. The Gift—Fueling Nonexclusive Global Development

What has transpired to date is admittedly just an initial portion of the vaccine R&D process. It is, however, a crucial portion—the one containing the basic ideas being explored.

We could consider imposing conditions of affordability and open IP on those who use the candidates developed, if feasible. At a minimum, the open display would constitute prior art, and may prevent certain items from being patented by other parties.

With the results being fully displayed and open—not just the data, but the IP itself, and the thinking that went into it – promising candidates are there for the world to take forward. And in times like these, the demand is immense. Additionally, the open, global process would sidestep the concerns of national boundaries, nationalism, IP exclusivity, public subsidization of private entities, and universal provision that bedevil the current model.

This is a project of the Open Source Pharma movement, as supported by the Open Source Pharma Foundation, www.ospfound.org, a global nonprofit.

An initial partner includes the European Vaccine Initiative http://www.euvaccine.eu/