Machine Learning-assisted Research on ME/CFS

mariovitali

Senior Member (Voting Rights)
Dear All,


I wanted to present here my Research that is performed using advanced analytical techniques. Many of you may know me from Phoenix Rising but i wanted to take the opportunity to present this work here as well.


Regarding my background , i am a Data Scientist / Machine Learning Engineer with 18 years of experience. You can see my LinkedIn profile here. I also began making tweets to specific people / organizations on twitter, hoping to turn their attention. My twitter account can be found here


Before continuing, it is important to say that this Research was able to identify key elements of ME/CFS (and possibly other syndromes such as Post-Finasteride Syndrome, Gulf War Illness Syndrome, etc) before any other Researcher.

These elements are the following :


-Pyruvate Dehydrogenase Complex

-Phospholipids Metabolism

-Bile Acids Metabolism

-Possible Liver Involvement


The work (a 32-page document) has been first presented to Professor Ron Davis in October 2017 who got interested after some findings and also to Professors Derya Unutmaz and Maureen Hanson who have found the hypothesis being discussed as very / quite interesting respectively.


I am also in talks with Professor Modra Murovska of the EUROMENE Network so that i may present this work this coming September in London.


In a nutshell : Machine Learning, Network Analysis, Natural Language Processing and Root-cause Analysis techniques were used to identify the most probable biological targets as origins of several syndromes, including ME/CFS. All techniques identify the Liver as the main area of interest.


The original post on Phoenix Rising can be found here :




Here are some latest findings using the techniques discussed :



- Primary target is still Liver function. A "Liver Stressor" such as a Virus, Medications or even prolonged Stress may disrupt Liver function which in turn disrupts key metabolic pathways that include Bile Acid Metabolism, Endoplasmic Reticulum Stress, Phagocytosis, Vitamin K Metabolism (list not inclusive)


-The importance of Gut, Gut Microbiome and Gut lining is confirmed by these techniques. However we need to ask why we have altered Gut Microbiome and impaired Gut Lining : Could impaired Liver function be responsible for this? An important paper for those looking at the Gut Microbiome can be found here :

https://www.nature.com/articles/s41575-018-0011-z

-The relevance of Sepsis, Lactate are also confirmed.



See below the ranking of Topic importance :

fs1.png


There are also some key questions that are to this day unanswered :



1) What is the prevalence of *any* of the following conditions to ME/CFS Patients ? :

https://bit.ly/2s3YzFy (Note that 3 tabs exist)

2) How many ME/CFS Patients have Liver Fibrosis ? (a test called Fibroscan may be used for this)


3) What is the percentage of ME/CFS having impaired Bile Acids Metabolism? Despite the extensive testing, a very simple test such as Total Bile Acids (TBA) has not been performed on a large scale to ME/CFS patients to this day.


My goal is to have Researchers look more closely to the Liver. In my e-mails to several of them i have stated that the following are readily available for evaluation under strictly controlled conditions where applicable :



1) A Number of SNPs with potentially high relevance to ME/CFS and several other syndromes

2) A Personalised Regimen that may ameliorate or even halt ME/CFS Symptoms

3) A methodology that is able to predict the severity / existence of Symptoms


Ultimately, i am looking to collect a large number of ME/CFS Patients with the Issues listed on the file i linked above and to convince Researchers to look at this Hypothesis.

I further hypothesise that ME/CFS Research has not advanced simply because ME/CFS Researchers have made the false assumption that since Liver Enzymes such as SGOT, SGPT, γ-GT are not elevated, therefore Liver function is not an issue. Unfortunately, the only way to fully assess Liver function is through Liver Biopsy from multiple sites.


Thank you for your attention.
 
Last edited:
FWIW, in the first few years of the illness, I had a couple of liver function test results that seemed to concern the doctors who ordered them. Each time, the tests were normal on retesting. Maybe that's par for the course with liver functions tests, or maybe it's a more frequent occurrence in ME patients. I don't know. The doctors would ask how much alcohol I drank, and they seemed somewhat incredulous when I told them that I never drank alcohol.
 
@strategist

I wish i could but the problem is that Liver Biopsies may have complications. Nevertheless what i am trying is to build awareness, first among patients. I would also like to present several cases with Liver issues to Professor Murovska of the EUROMENE Network. I have sent to her (and also Ron Davis, Robert Phair, Derua Unutmaz, Maureen Hanson and many others) Fibroscans of ME/CFS patients with Liver fibrosis F2 or F3 (this is quite significant level of fibrosis, especially F3)

We have many immunologists on board but i hypothesise that a very significant percentage of ME/CFS patients have some kind of Liver issues. How many patients have problems with Iron? How many have Wilson's Disease? The list of potential Liver issues is quite large.

Interestingly, many viruses can disrupt Liver function (one of them is EBV) but also read the following post on PR :

https://forums.phoenixrising.me/ind...e-thank-you-for-everything.60030/#post-979904

Basically, any Virus that has been found to affect Liver function can set the stage to ME/CFS on certain individuals according to my hypothesis. To be more precise, anything that can affect the Liver may be setting the stage for ME/CFS (even a Cholecystectomy).
 
FWIW, in the first few years of the illness, I had a couple of liver function test results that seemed to concern the doctors who ordered them. Each time, the tests were normal on retesting. Maybe that's par for the course with liver functions tests, or maybe it's a more frequent occurrence in ME patients. I don't know. The doctors would ask how much alcohol I drank, and they seemed somewhat incredulous when I told them that I never drank alcohol.
Same for me. Then the LFTs would became elevated again with crashes or relapses. For the last year or so they’ve remained slightly elevated and have been chalked up to presumed NAFLD.

@mariovitali thank you for sharing. Will be interested to see what comes of this.
 
I wish i could but the problem is that Liver Biopsies may have complications. Nevertheless what i am trying is to build awareness, first among patients. I would also like to present several cases with Liver issues to Professor Murovska of the EUROMENE Network. I have sent to her (and also Ron Davis, Robert Phair, Derua Unutmaz, Maureen Hanson and many others) Fibroscans of ME/CFS patients with Liver fibrosis F2 or F3 (this is quite significant level of fibrosis, especially F3)

It is possible to examine the liver of deceased patients. There is a group in the UK that is doing post mortem studies. Charles Shepherd should be able to tell you more.
 
@Trish

Yes, Thanks for pointing this out. What is also interesting are the latest findings of Maureen Hanson. In one of her papers, the following is stated :

Primary bile acids such as sulfoglycolithocholate (13) are synthesized in the liver and the major bile salts result from its conjugation with taurine (28) and glycine, forming taurochenodeoxycholate and glycochenodeoxycholate (21), respectively, which are all numbered metabolites found to be significantly reduced in ME/CFS patients. Reduction in these compounds is suggestive of damage to the liver. A study from the FDA National Center for Toxicological Research (NCTR) was able to identify liver injury biomarkers as the result of drug-induced hepatotoxicity in rats25. Strikingly, several other metabolites identical to our findings were also identified in their report, namely 5-guanidino-2-oxopentanoic acid (12, also named 2-oxoarginine), sebacic acid (60), along with energy metabolites from the glyoxylate and dicarboxylate metabolism. These biomarkers could be used to define a serum metabolic signature by creating a panel for hepatotoxicity prediction25. Whether ME/CFS patients actually have liver damage as a result of the disease is not known, as it also could be a consequence of medication toxicity in this patient group, which often uses prescription or over-the-counter pain relievers. Abnormally elevated levels of fatty acids 72, 73 and 74 (Table 2) could be a sign of hyperlipidemia and could also be related to a deficiency in liver activity.

Link : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5365380/

Some questions :

-Why are Bile acids reduced?
-Why do we have metabolites suggesting hepatotoxicity?
-Why many ME/CFS patients cannot tolerate any medication?


@strategist

This is very interesting, can i contact him?


Unfortunately it is not proper to post here the snapshots i have but from a post i made to a FB Group of ME/CFS Patients, many of them started writing how they got ME/CFS from Gallbladder removal or having Hepatitis, or Primary Biliary Cirrhosis (PBC) or Porphyria, having also Wilson's Disease, hemochromatosis etc. All of these may be affecting proper Liver function.

I cannot readily dismiss that Confirmation Bias may be involved here, however i do strongly believe that someone must look at enterohepatic circulation and Liver function ASAP.
 
@mariovitali I have low levels of albumin, transferrin, and ceruloplasmin. I note that these are all synthesised in the liver. Do you think these findings are relevant to your hypothesis? Are they common findings?

Also, while I rarely see it discussed my me/cfs doctor who has been working exclusively in me/cfs for several decades finds high ferritin (without high iron) as one of his most common findings...Do you think this might be liver related?
 
Some posts on this thread are straying beyond a general discussion of hypotheses about possible causes of ME. Please can members remember this forum rule:
Rule 5: No medical advice
Members may offer or request opinions and personal experience regarding seeking a diagnosis and treatment, but they may not diagnose or recommend treatments to each other, nor request such diagnoses or recommendations. Treatment recommendations are also inappropriate if contained in a protocol which purports to be broadly applicable based upon a diagnosis or other characteristic, or includes a process for determining which people should try specific aspects of the protocol.
 
I tested positive for antimitochondrial antibodies which had me sent off to a liver specialist to be checked for primary biliary cirrhosis a few years after I first became ill. After more tests they just said they couldn’t find anything but i’ve often wondered what was going on there and if my liver was something that should be looked at more closely.
 
@Diluted-biscuit

Please note that i am not a Medical Professional. I looked at Wikipedia about AMA and i read :

Anti-mitochondrial antibodies (AMA) are autoantibodies, consisting of immunoglobulins formed against mitochondria,[1] primarily the mitochondria in cells of the liver.

The presence of AMA in the blood or serum of a person may be indicative of the presence of, or the potential to develop, the autoimmune disease primary biliary cirrhosis(PBC; also known as primary biliary cholangitis). PBC causes scarring of liver tissue, confined primarily to the bile duct drainage system. AMA is present in about 95% of cases.[2] PBC is seen primarily in middle-aged women, and in those afflicted with other autoimmune diseases.

It is interesting that it is present in 95% of cases with PBC, perhaps @Jonathan Edwards would like to comment.
 
@Diluted-biscuit

Please note that i am not a Medical Professional. I looked at Wikipedia about AMA and i read :



It is interesting that it is present in 95% of cases with PBC, perhaps @Jonathan Edwards would like to comment.

I’ve been seen recently by a consultant and AMA weren’t present, nor did any other test results come back with problems. It’s just always something I’ve been curious about.
 
That’s correct, with about a 20 year gap in between.

It’s probably completely unrelated to your research so sorry to hijack your thread. The liver connection was just interesting.

*edited to add paragraph*

No problem at all. To be honest i am still very curious as to how this specific marker has been found to be positive and then changed.

Thank you for sharing this.
 
@mariovitali I have low levels of albumin, transferrin, and ceruloplasmin. I note that these are all synthesised in the liver. Do you think these findings are relevant to your hypothesis? Are they common findings?

Also, while I rarely see it discussed my me/cfs doctor who has been working exclusively in me/cfs for several decades finds high ferritin (without high iron) as one of his most common findings...Do you think this might be liver related?
Perhaps of interest - ferritin and GGT/ liver link

http://www.thefatemperor.com/blog/2...d-sibo-with-gut-based-root-cause-interesting-


http://healtheiron.publishpath.com/GGT-science-library
 
Last edited:
Just wanted to provide an update based on the latest results of the algorithmic approach i've been using.

Using Network Analysis, Osteopontin has been identified as a potential Research target. Regarding Osteopontin and its role to immune function we read from Wikipedia :

Role in immune functions[edit]
As discussed, OPN binds to several integrin receptors including α4β1, α9β1, and α9β4 expressed by leukocytes. These receptors have been well-established to function in cell adhesion, migration, and survival in these cells. Therefore, recent research efforts have focused on the role of OPN in mediating such responses.

Osteopontin (OPN) is expressed in a range of immune cells, including macrophages, neutrophils, dendritic cells, and T and B cells, with varying kinetics. OPN is reported to act as an immune modulator in a variety of manners.[6] Firstly, it has chemotactic properties, which promote cell recruitment to inflammatory sites. It also functions as an adhesion protein, involved in cell attachment and wound healing. In addition, OPN mediates cell activation and cytokine production, as well as promoting cell survival by regulating apoptosis.[6]


It is interesting to see also that there is a relationship of Osteopontin with MFGE8, a target previously identified by these methods.

More specifically :


MFG-E8 (milk fat globule-epidermal growth factor 8) deficiency is strongly associated with acquisition of immune-mediated disorders due to the loss of tissue homeostasis. However, comparatively little is known regarding its functions in gastrointestinal tract disorders, in which immune homeostasis is a major concern. Herein, we report altered MFG-E8 expression in inflamed colons during the acute phase of murine experimental colitis and found that treatment with recombinant MFG-E8, but not its arginine-glycine-aspartate mutant counterpart, ameliorated colitis by reducing inflammation and improving disease parameters. To reveal the MFG-E8-mediated antiinflammatory mechanism, we employed an in vitro system, which showed the down-regulation of NF-kappaB in an LPS-dependent manner. Additionally, MFG-E8 altered alpha(v)beta(3) integrin-mediated focal adhesion kinase phosphorylation by impeding the binding of one of its potent ligands osteopontin, which becomes activated during colitis. Taken together, our results indicated that MFG-E8 has a novel therapeutic potential for treatment of colitis.

Link : https://www.ncbi.nlm.nih.gov/pubmed/19454719
 
Back
Top Bottom