Phair came to this disease like many others have – he knew someone (a neighbor) who had it. When a piece on ME/CFS by Tracie White showed up in the Stanford Alumni magazine in 2016, he contacted Laurel Crosby in Ron Davis’s Stanford lab and has been volunteering ever since. He noted that, “Without that story, I wouldn’t be here.”
Without that story, we wouldn’t be talking about a new hypothesis for chronic fatigue syndrome (ME/CFS) – one which could – we’re a long way off from demonstrating this – but one which could, if it’s correct, offer a cheap, simple and quick solution to ME/CFS.
That’s the power of sharing. The more sharing – the more we spread the word about ME/CFS – the better chance we have of finding the person who produces the breakthrough that turns this disease around. Who knows? It could be Robert Phair…
Robert Phair had chanced on something that nagged at him. It was a metabolic result in the Severe ME/CFS Patient study that just didn’t make sense to him. Many of us would probably move on to what we did understand, but the disconnect really bugged Phair.
As Ron Davis later explained, it’s often the oddball results that intrigue good scientists. They’re more interested in what they don’t know than in what they do. When something doesn’t fit their conceptions, it bugs the heck out of them. They feel compelled to figure it out.
The saga of the ozone hole demonstrates how ignoring anomalous data can have severe consequences. Atmospheric researchers had been getting and deleting weird data from the upper atmosphere in their models for years. Because it didn’t fit their conceptions, they simply assumed it was an outlier and eliminated it. Doing so for much longer, they later realized, could have had enormous consequences – the data actually indicated that a hole in the ozone layer was forming.
When Phair dug into his strange data, it ended up opening a new world for him — and perhaps for us, too.
In a way, it was no surprise that Phair dug in the way he did: figuring out answers to complex biological questions is what he does for a living. Phair is an engineer and biologist who’s made a career out of understanding the functioning of complex systems. Phair’s company – Integrative Bioinformatics – uses a process called mechanistic kinetic modeling, which allows researchers to test complex hypotheses against experimental biomedical data. The mathematical descriptions of those systems that kinetic modeling creates allows researchers to assess the dynamics of those systems; i.e. to describe how they work and how they can be changed.
A “Metabolic Trap” Emerges
As Phair dug deeper into the possible causes of the weird data, a “metabolic trap” – a kind of biological sinkhole – opened up before his eyes. Once the process – which involved amino acid oxidation – started, he saw no way for an ME/CFS patient to get out of it without outside help (e.g. a treatment). Looking further, he and Davis realized it could conceivably explain some fundamental symptoms in ME/CFS.
Since then, he’s been creating model simulations to test his hypothesis. Thus far, he’s created kinetic models of the central metabolic systems in the body (mitochondrial electron transport chain, TCA cycle, fatty acid beta oxidation, amino acid oxidation, glycolysis and pentose phosphate pathway, purine synthesis and degradation, and NAD synthesis).
Using Davis’s genomics data from the Open Medicine Foundation’s Severely Ill Big Data Project, Phair has examined the exons (the part of the DNA which actually codes for proteins) of 100-plus genes involved in energy production. First, a variant commonly found in the general population — but which was more commonly found in ME/CFS — popped up. Digging deeper two more gene variants were found.
I asked Phair if he had finished up the genetic testing – the answer was far from it!
No. Most CFS investigators focus on rare SNVs. To me it’s just a matter of logic that predisposing genetic factors for ME/CFS must be common, not rare, in the population. Otherwise you can’t explain the many well-known CFS epidemics like Incline Village and Lyndonville. The downside of this logic is that common mutations are common! So, we still have many thousands to examine. Fortunately, we have colleagues at Stanford and at Harvard who are great at big data analytics.
There are two sides to modern systems biology – the mechanistic hypothesis testers like me, and the statistical hypothesis generators – like all the AI ads you see on your devices. We think these two groups of researchers should be working together on tough problems like ME/CFS.
Further analysis suggested that these gene variants may significantly, and in a negative manner, impact the functioning of key enzymes that process important metabolites involved in energy production, brain and immune functioning. When Phair added the impaired enzyme functioning to his metabolic modeling to determine if it could account for the strange metabolomic finding, he found that it could.
Testing the Hypothesis
That hurdle passed, it’s time to test the hypothesis experimentally — something, Davis noted, that Stanford has the tools to do. (Stanford was recently rated the third best medical research school in the country.) Davis and the OMF has made testing Phair’s hypothesis a major focus.
The team will use “tracer” metabolites to determine if the cells with these mutations are, in fact, functioning less effectively. Phair said they should know if the metabolic trap is present in the white blood cells of ME/CFS patients by the end of summer.
The potentially very big news regarding Phair’s metabolic trap is that, as noted above, if it works out – Ron cautioned that’s a big “if” at this point – it could conceivably lead to a cheap, fast and effective treatment. It might also be able to help explain how some autoimmune diseases occur.
Because the treatment strategy would involve tweaking one of the major systems in the body, though, it has the potential to do either good or harm. Worried that in the wrong hands, it could backfire – making ME/CFS patients worse – Davis and the Open Medicine Foundation are keeping the details of the possible metabolic trap under wraps until they know more about it.
Looking ahead, I asked Phair what’s next if the “metabolic trap” hypothesis passes the next hurdle?
If the metabolic trap can be demonstrated experimentally in cells from CFS patients and not in healthy controls, we’ll begin the search for ways to return such cells to normal. With a little luck, as Ron says, this might turn out to be the easy part.
The tougher question is what’s next if we cannot demonstrate the metabolic trap in these easily obtained blood cell types. People tend to think one cell is as good as another, and for genomics that’s true, but there are well over 300 different cell types in the human body, and it’s likely that only some of them are affected in ME/CFS. But which ones, and how can we get those cells to study?
Maybe we’ll have to move to stem cells. Our current thinking, though, is that we should be able to induce the metabolic trap in cells carrying the mutations we’ve identified. So that’s the first backup plan.
Oh, and it’s always possible the metabolic trap will turn out to be “a beautiful theory destroyed by an ugly fact.” I wake up many mornings thinking of another reason I might be wrong – then I try to find a way to test that possibility. It’s how science works.
Phair has put out a call for whole genomic data from ME/CFS patients.
Whole Genome Data
We are seeking CFS/ME patients who have had their genomes sequenced and who would be willing to share their data for research purposes.* Your data would help us test theories about gene variants that may have predisposed you to developing CFS/ME when you encountered the infection or other stress that was the cause of your case of CFS/ME. We cannot provide feedback or use the data in studies. Nevertheless, confirmation or refutation of what we find in professionally acquired and processed WGS or exome sequencing data could help us a lot in our search for something that helps every ME/CFS patient.
You can reach me at email@example.com. I am working on CFS as a volunteer because I have a friend with the disease and another friend whose son is severely ill with CFS. There is no charge for my work on CFS.
Your data would be most comprehensive and powerful in the form of .VCF (Variant Call Format) file but we will work to find a way to analyze your data in any format you are able to provide.
Please also tell us your age and gender and write a paragraph or a page with a description of how you came down with ME/CFS. A list of current symptoms is optional, but desirable.
*23andME, Ancestry.com do not do whole genome sequencing but could be helpful. Find out about WGS here. Companies that do it include Illumina, 10xGenomics, Veritas, and Qiagen. As of 2016 a whole genome sequence from Illumina or Veritas costs about $1,000 but Illumina hopes to bring the price down to $100 in the next 3-10 years.
Robert Phair on using 23andME/Ancestry.com
23andMe covers only 3 of the five variants that are important to the current metabolic trap hypothesis. So 23andMe data sometimes works and sometimes has insufficient information. The 23and Me file that I need has a filename like this: genome_firstName_lastName_Full_date_time_in_digits.txt
so if it was mine it would be something like
.zip instead of .txt is also fine
If you can open the file in a text editor it should begin with a header that looks like this:
# This data file generated by 23andMe at: Wed Jun 14 12:04:30 2017
# This file contains raw genotype data, including data that is not used in 23andMe reports.
# This data has undergone a general quality review however only a subset of markers have been
# individually validated for accuracy. As such, this data is suitable only for research,
# educational, and informational use and not for medical or other use.
# Below is a text version of your data. Fields are TAB-separated
# Each line corresponds to a single SNP. For each SNP, we provide its identifier
# (an rsid or an internal id), its location on the reference human genome, and the
# genotype call oriented with respect to the plus strand on the human reference sequence.
# We are using reference human assembly build 37 (also known as Annotation Release 104).
# Note that it is possible that data downloaded at different times may be different due to ongoing
# improvements in our ability to call genotypes. More information about these changes can be found at:
# More information on reference human assembly build 37 (aka Annotation Release 104):
# rsid chromosome position genotype
This will be followed by many thousands of lines each corresponding to a single genetic variant.
Ex-Weightlifters / Bodybuilders
Phair is interested in knowing the supplement intake of ex-weightlifters or bodybuilders as they came down with the disease. He’s also interested if anyone doing these activities was taking in any extreme amounts of one particular food or drink.
Regarding this question, Phair said he was asking more out of curiosity than necessity. For me, I’ve always been interested in the number of people who were workout nuts – as I was – who fell prey to ME/CFS. Phair said:
I have a hunch surrounding people who train hard and fall victim to ME/CFS, but I’m not in a position to test that hunch in patients. I’m in the gym 6 hours a week myself, so there’s a kind of camaraderie behind my question. The stories of athletes in films like Unrest contain just as many new clues as do the more frequent stories of the first crash.
To me understanding crashes is critical. One of the most compelling features of the metabolic trap is that it provides a straightforward explanation for the initial crash. I’ll read every response that patients might share on these issues, because you never know when unexplained details you’ve been storing for a long time will suddenly merge into a simple theory.
We cannot use data acquired this way in our publications, because we can’t vouch for the diagnosis or the sequencing pipeline. And two, we cannot provide feedback to individual patients who send us their whole genome VCF files. Nevertheless, confirmation or refutation of what we find in professionally acquired and processed WGS or exome sequencing data could help us a lot in our search for something that helps every ME/CFS patient.
Davis and the Open Medicine Foundation must be happy to see the Severely Ill Big Data Project already paying dividends: the “metabolic trap” hypothesis was birthed out of it.
Phair’s growing presence at Davis’s lab is notable for several reasons. Phair’s definitely got some skin in the game – he’s got a friend with ME/CFS – but it’s clear two years into his work on ME/CFS that intellectual excitement is a big driver. Having folks like Phair and the Cortene team find fertile ground in ME/CFS is good news indeed for a field that needs as many bright minds as possible.
On that note, sometime this year, Davis and the OMF will start making data from the Severe ME/CFS Big Data project available to outside researchers. We’ll see what other ideas it may spawn.