M.P. Janson Institute for Analytical Medicine

We are a small research institute applying advanced mathematical and computational techniques and cross-discipline knowledge to medical issues. We have the following areas of research:

• Topological Data Analysis of euthyroid homeostasis as a method of identifying subclinical thyroid dysfunction
  

  Using TDA analysis of the primary hormones in thryoid function, we hope to develop an N-dimensional "shape" that conveys healthy thyroid function. Current statistical analysis fails to consider being "low-normal" in opposing feedback loop mechanisms as an indication of an underlying issue. TDA offers a way to identify patients that may be outside of the aggregate homeostatic profile while being within "normal" test ranges (which are statistical, not stoichiometric, values).

  
  
• Vectorization of pattern matching in genomic profiles to identify molecular mimicry of pathogens in human tissue
  

  A great deal of research is focused on genetic auto-immune diseases. However, "acquired" auto-immune disorders mimic their genetic counterparts without the correlating antibodies. We believe this is the result of molecular mimicry in which pathogens use immuno-evasive strategies to fool the host's immune system. We are actively developing tools to use high-performance vector hardware to look for off-axis nucleotide matches between pathogens and hosts, and "proof-of-concept" efforts have already yielded interesting results. Have symptoms of Lyme disease but ambiguous B. burgdorferi test results? Get tested for B. miyamotoi.

  
  
• Functional Integration as a basis for time-dependent spatial dispersion of a pathogen in a community
  

  Functional integration is the basis for the solution to the continuous random walk problem, and the mathematics underpenning Richard Feynman's alternative derivation of quantum mechanics. Each path taken contributes a probability towards the final calculation. A pathogen can take a nearly random "walk" through a community and undergo both cascading and absorbing events, where the instance of the pathogen that arrives at Point B is not the one that left Point A. The discrete solution is known as "Transitive Cascading Markov Chains," and has been effectively used to model power grid failures.

  
  
• Metabolic dysfunction due to CFS/CAEBV
  

  There is a clear subset of Chronic Fatigue Syndrome (CFS) patients that have Chronically Active Epstein-Barr Virus (CAEBV), the same virus that causes infectious mononucleosis and infects more than 90% of the world's population. It can reactivate later in life, much like its more commonly known family members Shingles and Herpes Simplex, without the clearly visible symptoms. The EBV virus has a high affinity for lymphatic tissue and is often not readily detectable in the blood, leading to missed diagnosis. Our research has found that there are known mechanisms explaining the fatigue, better testing protocols than are commonly used, and possible treatments. It might not even be a matter of strengthening your immune system; you have to get the virus to stop reproducing, too.

  
  
• Pharmacogenomics for better efficacy of medications
  

  We are not all horses. Many medicines lacked genetic diversity in the test populations, leading to medications that work great for some people but have adverse effects for others. This particularly includes members of minority groups, which can lead to (additional) mistrust of medical science. Pharmacogenomics, the analysis of genetic factors in the production of enzymes that are utilized to convert prodrugs to drugs and drugs to metabolites, offers an explanation for the occurrences of side effects as well as approaches for the patient to educate their doctor on the best medications for their conditions. You might rapidly metabolize the prodrug but poorly metabolize the drug, leading to high doses in a short period of time you can't clear.

  
  
• Carbon footprint for computing
  

  Because of Germany's strong laws on data protection, our email and web server is hosted there. However, we do all of our computing in-house in the U.S., and we are keenly aware of the energy consumption this requires. A fair amount of our efforts focus on low-energy, high-performance computing, where we write better algorithms instead of throwing bigger hardware at the problem. As we move to solar power for our energy source, we like to say "Ah, it's a good day - not a cloud in sight."

  
  

We have found that reductionism in medical science is a tremendous impediment to advancing medical care of patients. We advocate and develop cross-discipline methodology that incorporates frontline research in medical science, community-based patient-centered medical care, and advanced mathematical and data-analytical techniques to provide scientifically-rigorous approaches to difficult medical problems.

Almost all of our analytical tools are developed in-house, and we host our own hardware. We do not store data on other people's computers (otherwise known as "the cloud") nor do we utilize tools that we can't verify ourselves. If you believe science advances by never being wrong, we are not the place for you.

If you want to learn more or participate in our efforts, please contact us at IAM at this domain (mpjanson) dot org.