Authored by Jon Fleetwood, via Substack,
A sprawling, multi-institution effort for “predicting” future pathogen characteristics, developing vaccines “in advance of need,” and building the very systems intended to validate those predictions.
The Defense Advanced Research Projects Agency (DARPA) spent the early 2010s constructing what may have been the most ambitious predictive vaccine-development infrastructure ever attempted: a sprawling, multi-institution effort designed to determine the future characteristics of purported pathogens before they emerged and ultimately use those predictions to develop drugs and vaccines before they were needed.
The program, known as PROPHECY—short for Pathogen Defeat—was announced in 2010 under Broad Agency Announcement DARPA-BAA-10-93 and was managed by DARPA’s Defense Sciences Office.
DARPA itself described the vaccine-centric purpose of the program unambiguously.
According to the agency: “The Prophecy (Pathogen Defeat) program will explore the evolution of viruses in the hopes of predicting viral mutations and ultimately developing drugs and vaccines in advance of need.”
The effort was not limited to coronaviruses, influenza, or any other single disease category.
DARPA repeatedly stated that the goal was understanding: “the natural evolution of any virus.”
The result was a massive architecture that brought together machine learning researchers, statisticians, bioinformaticians, computational biologists, laboratory scientists, surveillance specialists, universities, contractors, and national laboratories into a single “predictive” framework.
Yesterday, this website reported that DARPA’s PROPHECY program expanded into the laboratory of coronavirus researcher Ralph Baric years before the COVID-19 pandemic and nearly a decade before the DARPA/DEFUSE proposal documented all three defining structural features of the SARS-CoV-2 spike protein prior to the outbreak.
DARPA Wanted to Turn Vaccine Development from Reactive to Predictive
DARPA openly believed that existing vaccine development operated backwards.
The agency described the modern pharmaceutical model as: “observational and reactive.”
PROPHECY was intended to replace that model with one that was: “predictive and preemptive.”
Rather than waiting for future purported disease threats to appear before beginning the process of characterization and countermeasure development, DARPA sought systems capable of determining the characteristics of future pathogen populations in advance.
Everything in PROPHECY flowed from that objective.
Building the Blueprint Before the Pathogen
DARPA’s ambitions extended far beyond outbreak forecasting.
The agency stated: “DARPA seeks to achieve the ability to successfully predict the natural evolution of any virus, via platforms and algorithms which are capable of monitoring rare advantageous viral events…”
The program sought systems capable of determining:
- future mutations,
- future reassortments,
- future genetic events,
- the order in which mutations would emerge,
- genotype-to-phenotype relationships,
- and ultimately the characteristics of future pathogen populations.
DARPA acknowledged: “there is at present no reliable capability to predict viral reassortment or mutations responsible for the emergence of new viral strains.”
Its proposed solution was: “An investigative platform to predict mutations and possibly reassortments in advance of their occurrence.”
The agency further explained that PROPHECY sought: “predictive algorithms of viral evolution that are informed and validated experimentally using high throughput biological platforms…” and: “the integration of a viral evolution platform with an algorithm that predicts mutations which confer an evolutionary advantage to the viral population.”
Among the formal requirements of the program was the ability to: “reproducibly predict the genomic, proteomic, and/or functional attributes of the final viral population.”
DARPA also sought to determine: “the rate and order of mutation acquisition” as well as: “the correlation of genotype to phenotype.”
The objective was not merely individual mutations.
It was the endpoint population itself.
Prediction Was Only the First Layer
PROPHECY was never designed to stop at prediction.
DARPA required participants to construct what it called: “biological validation systems” and repeatedly emphasized: “testing and validation of the system and algorithm.”
The BAA instructed performers to: “Perform a ‘real world’ test of the predictive algorithm and biological validation system…”
The purpose of that exercise was to determine whether researchers could: “reproducibly predict the genomic, proteomic, and/or functional attributes of the ending viral population.”
DARPA further instructed participants to: “develop a research strategy and solicit collaborations from partners” that would assist in: “applying their predictive methodology to surveillance data from the real world.”
The architecture described by PROPHECY therefore extended beyond prediction alone.
The program sought to construct systems capable not only of generating future pathogen blueprints, but also of attempting to validate those predictions experimentally and through subsequent observational frameworks.
Bottom Line
Amid COVID-19 debates, public attention has focused on later programs such as DEFUSE, ADEPT, pandemic mRNA platforms, and Operation Warp Speed.
But PROPHECY confirms that by 2010, DARPA was openly attempting to build a generalized predictive infrastructure capable of deciding the future characteristics of purported pathogens, using those predictions to enable drugs and vaccines in advance, and constructing systems intended to validate those predictions experimentally and through later observational frameworks.
Most importantly, the effort was never intended for a single disease.
PROPHECY sought to build this capability for: “any virus.”
Whether viewed as predictive biodefense, anticipatory genomics, preemptive countermeasure development, or something else, the architecture described by DARPA followed a remarkably consistent sequence:
- vaccines as the objective;
- future pathogen blueprints as the enabling technology;
- “validation” systems as the final layer.
That architecture may ultimately prove to be one of the most consequential and least understood developments in the history of modern biodefense research.
