Thread by @Ayjchan

Starting my 🧵 discussion of gain-of-function research based on yesterday's twitter survey.

This might be incredibly long so I will be using gifs and graphics to help keep people awake on a Saturday morning.

First things first. I made the survey yesterday morning to get a sense of the public perception of "gain of function" (GOF) research.

This phrase has exploded in the media, even making its way into a congressional hearing.

c-span.org/video/?c496233…

It is clear that the public needs to know what GOF research means.

What does it mean when people say that the US might have funded GOF research at the Wuhan Institute of Virology?

In other words, the survey was not intended to get the specific opinion of experts who have spent the last decade (and more) debating gain of function research of concern (GOFROC) regulation (sorry, @R_H_Ebright).

As a newcomer to this topic, it was a slow process of realizing that my understanding of GOF - even as a scientist - didn't match the federal definition of GOF research of concern (GOFROC).

And that the federal definition of GOFROC leaves a lot of wiggle room for interpretation.

What I can gather, at the most basic level, what the public seems to perceive as GOF is that:

(1) A dangerous human pathogen has been derived.

(2) It was derived via unnatural processes, whether this included any level of genetic engineering or just selection experiments.

However, when we hear scientists throwing the term GOF back and forth, they are (hopefully) referring to the federal definition.

For the purposes of this thread, I'll rely on the most recent 2017 framework.

It's only 5 pages long but can be technical!
phe.gov/s3/dualuse/doc…

In this framework, although the main text doesn't specifically discuss GOF, it is aimed at scoping and guiding "HHS funding decisions on individual proposed research that is reasonably anticipated to create, transfer, or use enhanced [potential pandemic pathogens] PPPs."

What are potential pandemic pathogens (PPPs)?

These were defined to be likely highly transmissible AND likely highly virulent (causing significant disease) in humans.

We will see the use of "likely" come back to bite us later.

But when we talk about GOF, we're not talking about work with natural pathogens, even if they're dangerous PPPs.

We're talking specifically about enhanced PPPs aka PPPs resulting from artificial enhancement of pathogen transmissibility or virulence.

That means that even if someone goes out and collects hundreds of animal or human pathogens and brings them back to the lab, as long as there is no enhancement of the pathogen, it doesn't get covered in the discussion of what is GOF.

Another nuanced consideration is that artificially re-creating viruses found in nature or already infecting human populations is not considered as creating enhanced PPPs.

Eg, if I turn the original SARS2 into the new Delta variant of SARS2 in the lab, this is not considered GOF.

So, based on the federal definition, GOFROC is any research that enhances a pathogen such that it crosses the boundary and results in a PPP with both high transmissibility and high virulence in humans.

No matter how small the nudge.

In contrast, research that does not result in an enhanced PPP that is likely both highly transmissible and virulent in humans does not get classified as GOF(ROC).

Even if it involves copious amounts of genetic engineering.

Based on these visuals, I hope readers can already come to the realization that where that threshold line (shown in green) is drawn separating regular animal/human pathogens from PPPs is incredibly important.

This is where a lot of scientists get into fights with each other.

To make matters more contentious, the framework has this special caveat.

That's right. No matter where the research falls on the chart, if it was a modified PPP from surveillance activities (e.g., virus hunting) or vaccine development, then it isn't counted as an enhanced PPP.

Again, I keep having to say this, I'm not saying that this definition or framework is right or wrong, or that it even matches my perception of what should be considered GOF.

I'm saying it is what it is. If we don't like it, let's change it asap.

I really want to bring this home for the public.

Based on a scientist's individual judgement of what is likely to be a both highly transmissible and highly virulent human pathogen, the bracket of research counted as GOF(ROC) can vary quite a bit.

It's like trying to decide (for the first time) where the end zone in football should be.

And unfortunately, in this case, it's not that easy to draw one straight line down to separate GOF from non-GOF research.

Resulting in scientists basically doing this for years:

The same experiments can be deemed as GOF or non-GOF depending on which scientist you ask and their specific assessment of which pathogens are ***likely*** to be both highly transmissible and highly virulent in humans.

In addition, many have pointed out that the GOF debate distracts from the point that a PPP does not need to be enhanced in order to wreak havoc.

A 100% natural PPP collected and studied in the lab can also escape.

Many lab leaks have been of natural PPPs, not enhanced PPPs.

I hope this primer on GOF was useful. Going to take a short intermission and grab breakfast. I'll be back to discuss the poll results!

Meanwhile, something to ruminate: take a look at the EcoHealth Alliance x Wuhan Institute of Virology research proposal that was funded by the NIH and determine whether or not it is GOF or involved enhanced PPPs according to the 2017 framework:

grantome.com/grant/NIH/R01-…

The reason why I wanted to understand (and reveal) what the public thinks is considered GOF research is because people keep asking about my views on GOF.

I hope, from the primer, that it's clear that the very specific GOF federal definition is not what many people think GOF is.

Navigating GOF questions from journalists is still a minefield.

What if listeners or readers misinterpret that I think a whole bunch of virology should be banned?

Alternatively, what if they misinterpret that I don't think very dangerous GOF research should be banned?

Let's look at the poll results to consider the challenges of answering the question: What is GOF?

The scenario in Q1: intentionally enhancing virus infectivity or virulence is what a lot of people think is GOF, regardless of whether the virus is a PPP.

However, in Q2, once you remove the intent, about half of the people who thought a similar experiment was GOF changed their minds.

Some labs do this: passage natural viruses, trying different host cell types, to find a version they can grow well and study in the lab.

Serially passaging a virus can be reasonably anticipated to confer a GOF in the sense of enhancing its ability to infect and replicate in primate or human cells, for example.

But changing the intent (and toning down the language) made a large % of people change their minds.

The poll also revealed that the public, due to no fault of their own, is not well-informed about practices and experiments necessary for basic virology.

About 1/3 people think engineering cells to grow viruses is GOF. And about 1/8 people don't know.

Engineering cells to grow viruses can actually prevent against risky pathogen research.

For example, as opposed to serial passaging a virus until you get one that grows well on human cells, you've created a permissive host cell that doesn't select for new infection abilities.

Q4 also revealed a similar issue. That 1/5 people think creating animal models of human disease is GOF, and another 1/10 don't know.

The real problem is, I think, a lack of science communication surrounding what virologists have to do in the lab in order to understand how pathogens make us ill and how we can develop treatments against these infections:

The next pair of questions reflects a different issue in the GOF debate. What if, instead of doing GOF on actual viruses, I do it on a pseudotyped virus or a protein.

In other words, I obtain knowledge of how a virus can escape from vaccines, without creating the escape virus.

The majority (52%) of participants agreed that doing the experiment with the virus itself, creating potential escape variants, is GOF.

But if the experiment doesn't create these escape viruses, only 19% of participants still think it is GOF. In fact, 61% said it is not GOF.

But there are some scientists who even warn against the experiments not using live virus, because they fear that nefarious boffins will take this knowledge and do actual GOF work on pathogens - creating variants that escape vaccines and drugs.

That leads us to the next 2 questions: what if a feature is engineered into a virus based on knowledge that it will likely result in GOF?

Vast majority (92%) say this is GOF. But if you don't use live virus then 43% say it is not GOF.

The difficulty of calling GOF gets worse when you think about just switching parts between known human pathogens, instead of directly engineering in known GOF features.

And I thought it would've been worse still if you don't even know if the viruses you're working with are capable of infecting humans.

Surprisingly, the % of people who said this is GOF went up from 57 to 71%!

Thankfully, it went back down in the next question where both viruses being mixed-&-matched have not been shown to cause human disease.

45% people said this is not GOF.

I thought this next question would've resulted in more "I don't know"s.

Unfortunately, there are quite a few types of experiments that, regardless of intent, result in co-infection of cells or animals with multiple viruses.

For example, something as simple as inoculating an animal or cell culture with a sample from bat butts in an attempt to isolate novel viruses could result in multiple viruses infecting the same host.

Almost at the end!
This pair of questions again reveals that changing the intent of the experiment can change what people perceive as GOF.

Even though the experiment stays the same.
Even though the pathogen is not even enhanced.

I wrote this last question to stimulate some thought that viruses can be broken down into different functional parts.

Scientists can take these different parts and use them for good and/or for evil. It's not always clear cut even if it involves the same part being adapted.

Again, the poll is just for fun. Please relax, everyone, you're not being judged or graded. I can't see who answered which question and how.

If someone wants to make this a real public poll, it looks like there are a lot of people with useful feedback in these threads ;)

The point of the poll also isn't to mock the public. Like I said, even scientists have a hard time agreeing on what is GOF. Many don't agree with the federal definition+framework or how it is implemented.

Best thing you can do is ask each person: what do you think is GOF?