The quote below has been circulated for a while, amplified by those with no PCR understanding. Its focus is to make you believe that too many PCR cycles create false positives, directed mainly towards testing during the COVID-19 pandemic. Firstly, this “quote” doesn’t seem to exist as a real quote from Dr Kary Mullis. Secondly, the real part is really out of context. Thirdly, total PCR cycles aren’t the same as threshold cycles. Let’s step through this fake quote and lay out some realities.
These images are supposed to be “quotes” attributed to Dr Mullis. But the quoted part doesn’t seem to exist anywhere – not even at the site, fact-checkers say it came from.[9] In other words, this is not a quote from Mullis.[1] The meme quote images can be found by searching Twitter for “kary mullis, anyone” and wading through the resulting images.
Part of that quote is from Mullis, but the rest seems to have been bolted on to suit a narrative about too many PCR cycles producing false positive results during the COVID-19 pandemic. Which is untrue. What is true is that weak PCR positives (detected at later threshold cycles [CT], sometimes called ‘late positives’ in PCR parlance) may signal the very beginning or end of an infection’s course, or they may signify low levels of persistent virus ticking over. They don’t prove that a replicating virus was detected, but the presence of a replicating virus can often be safely implied in order to produce PCR positivity. For different reasons – each of these can be important to know and generally won’t be detected using virus culture or rapid antigen methods because they are too insensitive.
Let’s examine what Mullis did say and check if his words supported his stated intention.
Up front, Mullis was a Nobel laureate and a famous HIV-AIDS denier. For a more accurate summary of the early HIV story, you can read more accurate accounts about discovery [2] and some backstories [3,4]. Mullis was also a noted user of drugs, including LSD, so consideration should be given to the potential for that to impact his statements and views of the universe at any given time.[4,5,6]
The not-a-real-quote and the video it didn’t come from
The real part of the fake quote comes from a discussion panel Mullis attended, which was recorded. It is usually a clip of a longer video that you can view in its entirety, split into two parts here:
- https://archive.org/details/corporate-greed-aids-part-1
- https://archive.org/details/corporate-greed-aids-part-2
The content of interest starts around 48:36 in Part 2. But you won’t find that “quote” there, as it doesn’t exist; he doesn’t mention the first part, or the “cycles” or “you’re sick” bits, but he does say…
I don’t dispute the first part of that sentence, but please focus on his use of almost.
Sure, PCR can find almost anything, but only if the thing you built the PCR to detect is there to begin with. And if you built a good PCR, to begin with, of course.
PCR – especially the more modern version in use today – can’t find something that isn’t there or is present in amounts too low for that PCR’s dectection capability. Also, if the PCR is well built, it won’t detect things that are there but that its primers and probe aren’t designed to interact with (read more about real-time PCR).
Mullis, who won the Nobel Prize for his invention of an early form of PCR [7], also said, “I don’t think you can misuse PCR.” “It’s [PCR] a really quantitative thing. It tells you something about nature and about what’s there.” “That’s not a misuse it’s sort of a misinterpretation“. That last word, misinterpretation, seems very relevant since people shopping the fake meme seem to be misinterpreting Mullis’s words and PCR as it’s used today – a quarter century after this 1997 panel discussion and 40 years after Mullis published his relatively crude PCR method in 1983.
Ironically, in the video, Mullis started off by explaining how reality is so easily skewed by humans who miss the bigger picture (after having just chewed the leaves of a nearby tree).
He then went on to discuss a few skewed realities around drugs, HIV and some comments that helped provide fuel for the conspiracy theory I’m interested in here – that PCR is always positive “if you just run it for enough cycles”. Mind you, it’s a theory propagated mainly by those who haven’t worked expertly with PCR.
“Most of what the people that have been talking to you about today don’t have any goddamn thing to do with it”, said Mullis in the video. I’d have to agree, although I’ve taken that comment out of context 😁.
“This [re: HIV and AIDS] is not a little issue about laboratory testing”, “It’s a really big issue”. He went on to reinforce, at length, his ‘HIV/AIDS hypothesis is one hell of a mistake’ position, which so much data shows, was as wrong then as it is now. It has been argued that scientists with such minority views should refrain from campaigning for them.[8] But that wasn’t Mullis. Back to the video.
He rightly went on to say that “Science doesn’t work by opinions” and that “if I say ‘this is true‘, and I know it’s true because I’ve done experiments, then I have to be able to explain these experiments to any curious interested person.” Very true. And since he didn’t claim that PCR would always detect its target if it was just run for enough cycles, the underlying conspiracy must have been spun off by other actors using his quote as a starting point and to try add attribute credibility to it.
Mullis also stressed the need to adjust one’s thinking in the face of new data when he told the gathering, “Your final take on things can change every time you get new experimental data.” Perhaps Mullis was applying this approach when he said, “You can find almost anything in anybody.” Maybe he chose that word almost very deliberately, and whoever first concocted the meme quote has, in fact, done him a significant disservice. Okay, I’m being generous.
If we run a PCR for >40 cycles, is it always positive?
Hell no!
Firstly, can I remind you that each PCR is designed to detect a bit of the genetic sequence of the thing you are looking for? Do you honestly think you have the Ebola virus in you right now? What about Naegleria fowleri? In most cases, the answer will be ‘absolutely not’. Some of you will – sorry about that. But if I ran a PCR or an RT-PCR on the appropriate specimen – most of you would test negative because you simply don’t have a molecule of everything in your body all the time. That is an example where Mullis may have been giving us his unsupported opinion.
Let’s test this using a more reasonable target; human parainfluenza virus 4 (HPIV-4). It’s not a super common virus, but it does circulate. Below is an experiment from little ol’ me a decade or more ago (you can read more about this example here). I’m using this to specifically refute the opinion that PCR will always be positive for any molecule if you run it for more than 30 or 40 cycles (whichever version of the ‘too many cycles’ chant you may have read).
The two negative controls (water instead of patient sample extract) were PCR-negative (=target not detected). There’s no trace of an HPIV-4-positive fluorescent signal in the emission channel used to detect activity from this test’s fluorgenic probe. Huh? No target nucleic acids were detectable by PCR (55 cycles—which is a lot!). Nothing except the positive control produced fluorescence to indicate PCR amplification of the target molecules that crossed above the arbitrary threshold fluorescence value (shown by the dashed orange line) at a CT of about 30. Sorry, but it looks a lot like we don’t have a molecule of everything inside us.
In other words, modern real-time RT-PCR doesn’t detect something that isn’t there or in too low an amount for even PCR to detect. Yes, even very sensitive PCR has a limit of detection.
PCR can be contaminated – quality controls are essential.
I mentioned negative and positive controls above. Even though it was the ‘good old days’ wild west of PCR, it is sometimes difficult to find true negative controls among some of Mullis’s work. To be very clear, they are an essential inclusion for every PCR run. In my world, the absence of a no-template (water) negative control invalidates the entire PCR run.
I wonder whether Mullis was using too few negative controls and had some bad experiences that led him to believe you could “find almost anything in anybody.” I guess I’ll never know.
Is it even possible for PCR to be negative even after lots of cycles?
Hell yes!
The easiest way to know this is to look at one of the simplest pieces of evidence that repudiates those who misunderstood Mullis’s “you can find almost anything in anybody” using PCR. There are many people who get an HIV PCR test and test negative for HIV.
Another one? You probably had one or more COVID tests during the pandemic (a real-time PCR that tested for the SARS-CoV-2 virus), and your results were negative.
Mullis knew that PCR wasn’t always positive because he worked with it and published studies with negative results.
This fake quote really doesn’t stand up to scrutiny.
Mullis put his name to scientific articles and patents that refute the fake quote
In “The Unusual Origin of the Polymerase Chain Reaction,” sole author Mullis was clear that the PCR process starts “with a single molecule.” That implies it won’t start if no target is present. He didn’t qualify that with cycle numbers at all.
He spelled out that PCR also doesn’t need a pure biological sample of the thing your PCR is designed to detect. PCR is so specific it can identify its target from a mixture of other DNA and proteins. Those are his words, not mine![10]
Adding further to the irony of the fake meme, Mullis published a few peer-reviewed research papers, including one on the use of this earlier crude form of PCR for HIV DNA detection, showing that PCR was specific to HIV and not to other human retroviruses, HTLV-I or -II.
He knew the target had to be present to be detected. If it wasn’t there, it couldn’t be amplified.
Mullis also had his name on a number of patents, including one called “Detection of viruses by amplification and hybridization”, all involving PCR.[11,12]
It was exquisitely clear that a fairly crude version of the PCR was intended to detect all manner of targets “if present,” including microorganisms, whether contained in a pure or mixed sample.
At least while putting his name to these, he was clear on how specific he knew PCR to be. He knew it could identify what it was designed for, even if other stuff was muddying the waters. And he knew it could be negative, even if extra cycles were added.
Mullis showed the importance of a well-optimised PCR
Because Mullis’s most well-known works were in the earliest days of PCR – a lot was still being learned. Some of his papers excellently make clear how important it is to optimise a PCR before using it for anything important.
This is the kind of work done today by the big companies that supply PCR kits for pathogen detection and by any reputable and expert research team or pathology laboratory, which develops its own test from the ground up because there is not yet any good commercial option. The latter are the sources of early tests during an outbreak, epidemic or pandemic of a novel pathogen.
Optimisation involves choosing or fine-tuning the chemical constituents used in the PCR/RT-PCR recipe, storage life, freeze-thaw survival, the number of cycles and the temperatures used in each step. A badly optimised PCR will give bad results—a ‘garbage in, garbage out’ kind of thing.
Did Mullis develop the PCR methods we use today?
No.
Mullis didn’t develop real-time PCR; the method used most frequently today for virus, bacteria, fungi and parasite detection.
It’s the form of PCR that produces a fluorescent signal if there’s exponential amplification of the target and the signal accumulates in real time.
It’s fast, sensitive, and more specific because the probe provides an additional layer of sequence specificity.
Importantly, it’s a closed-tube method. This means that once you add your sample, you don’t have to open the tube again to get a signal as you did with earlier non-probe PCR methods and the early probe-based techniques in Mullis’ day. Opening the tube after the PCR risks contamination of the lab environment due to a positive reaction tube potentially containing trillions of copies of target DNA (hence the need for negative controls to watch out for that). We also use a few better components to make our PCR mixes than were available 30 years ago.
So, the PCR we use today is different in many ways from what Mullis was familiar with. Its use in a pathology setting is conducted with an order of magnitude more quality than you’d see in many experimental research settings.
Summary
It doesn’t look like this “quote”, often used to link Mullis to saying too many cycles caused problems with PCR during the COVID-19 pandemic, was a quote. His personal opinion that everybody has a molecule of everything in them isn’t supported by PCR. Mullis came from a different era of PCR, and things have changed a lot since those early days but even so, he put his name to publications that don’t support the intent of this fake meme. He also wasn’t working in a pathology setting, and the quality that is added to routine human testing is often quite foreign to researchers using PCR.
References
- https://fullfact.org/online/kary-mullis-anyone-PCR-test/
- https://www.nejm.org/doi/full/10.1056/NEJMp038194
- https://www.mayoclinicproceedings.org/article/S0025-6196(11)61990-3/fulltext
- https://www.iflscience.com/lsd-dna-pcr-the-strange-origins-of-a-biology-revolution-63126
- https://alumni.berkeley.edu/california-magazine/winter-2019/intolerable-genius-berkeleys-most-controversial-nobel-laureate
- https://boomcalifornia.org/2015/11/12/on-pcr-lsd-and-science-as-a-wild-ride/
- https://www.nobelprize.org/prizes/chemistry/1993/mullis/biographical/
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1757140/
- https://www.usatoday.com/story/news/factcheck/2022/01/14/fact-check-kary-mullis-quote-pcr-tests-outdated-lacks-context/9198197002/
- https://pubmed.ncbi.nlm.nih.gov/2315679/
- https://patents.justia.com/inventor/kary-b-mullis
- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1360491/
Views: 170