My professional involvement with biology started in 1973 when an abstract mathematical structure (the hyperoctahedral group, actually) which I had been studying made an appearance in an article in a mathematical biology journal. I was obviously curious.
From reading the article, I didn't believe that my expertise would help with the approach taken there at all, but I thought that there was an approach that would be fruitful.
I ran off to the nearby Jackson Lab with my idea and early results, and received encouragement – especially from two scientists, Drs. Don Bailey and Larry Mobraaten. They recognized an intersection of my approach with a symbolic but non-mathematical approach of Jan Hirschfeld, a serologist in Sweden, and put me in contact with him. I also received encouragement from the National Cancer Institute of the National Institutes of Health (NIH) with a grant that began in 1976 – and would continue with renewals until 1991. I was looking for logically consistent symbolizations of immunogenetic data – and I found them.
A 1987 meeting on the application of computers to biology featured the offerings of several companies. One application that impressed me was "cross-eyed stereo". By splitting the computer screen into a right half and left half, looking at the left half with one's right eye, and vice versa, one could see, in 3 dimensions, a visual computer representation of some molecule of interest. One could turn knobs, and so on, to take a visual tour through the molecule.
I was there to explain a computer program I had written that could help in finding the things I was after – most recently, the genes connected with the human major histocompatibility complex HLA. Dr. Devendra Dubey from the Dana-Farber Cancer Institute had provided me with HLA data – from real patients – and the program had suggested a model.
The standard model had been arrived at by first finding a gene that coded for an antigen recognized by an antibody in a person not having that gene. The gene was given a name, say A1, and the (minimal) reagent containing the antibody was called anti-A1. The process was then repeated, to find another gene suspect, and when it was shown that this was indeed new, it was called A2, and its defining reagent anti-A2. And the process was continued.
The process resulted in a standard model that had genes at three different locations (loci) on a chromosome: the A, B, and C loci. There was a one-to-one correspondence between the symbols for the genes and the symbols for the reagents defining them. The one-to-one correspondence was an immunological oversimplification, however. Everyone knew that there was cross reactivity. It was considered an annoyance, but not integral to obtaining a model for the immunogenetic system.
Cross reactivity, by the way, was the way the nature of immunity was first discovered: people that had had cow pox didn't get small pox. The antibodies that cow pox elicited – to fight the cow pox bug – were effective against the small pox bug.
One of the things that I had previously shown was that if the genes in a system were discovered in a different order (using exactly the same data on reactions), one would get different models for the system. One of the NIH reviewers for my grant renewals called this "the most important finding to come out of his work". But if different sequences of discovering the same facts lead to different conclusions, how confident can we be that one of them (the one history has presented us with) is correct. What we ought to see is things converging to the same model, whatever the sequence of discovery.
There are two things combining in an immunogenetic system: immunology and genetics. If we oversimplify things immunologically (by denying cross reactivity with our choice of symbolism), then the complexity of nature, not to be denied, shows up as a genetic over complication. The most noticeable of these over complications is linkage disequilibrium; that is, things don't turn out statistically as predicted.
In a previous book series contribution with Dr. Hirschfeld we had shown that linkage disequilibrium completely disappeared when the system (the human Ag blood group system) was modeled by the labeled reaction matrices. Dr. Robert Rosen, the editor of the series wrote us that our contribution was exactly the sort of intellectually stimulating material that he wanted in his series – more encouragement.
The paper with the new model for HLA was submitted to the Journal of Theoretical Biology. The reviewer abandoned his anonymity and sent me a copy of his recommendation to the editor: "His result on HLA is extremely important, and is likely to enhance theoretical biology in the eyes of experimentalists." More encouragement.
After my talk at the computers-in-biology conference, a man working for one of the computer companies came up to me. He said he was a former professor of chemistry, and he had made what to him was an exciting discovery. It seemed to promise opening up an entirely new field of chemical research. He published his paper on his discovery, and excitedly awaited the results. As for the results, his words were, "The silence was deafening." He said he had to tell me about this because he knew from my talk that same thing was going to happen to me.
He was right. There were experiments that could be done to confirm one model over the other. No one evidently wanted to do them. Theoretical biology was not much enhanced in the eyes of the experimentalists in HLA research, who as far as I know, continued using the old symbolism with all its oddities: for example, statistically expected gene combinations that never occur in the data – "perhaps" because the combination causes some deadly disease that kills the people that have it.
I never applied for a renewal of my grant beyond 1991. If people weren't interested in my results on HLA – the most important immunogenetic system in humans – there was nothing else I, a mathematician, could say or do.
I continued my interest in the works of others on the relation of diet/nutrition to disease. I knew I needed to listen to the pioneers in this area – and not the establishment.
Of course there were many experimentalists that had encouraged my work. At the Jackson Lab, after my work on the Ag system, but before working on HLA, the director came up after a talk I had given and told me, "This talk really turned me on."
I subsequently looked at the Jackson Lab's data on the mouse major histocompatibility system H2 (analogous to the human HLA system). It turned out that my computations (before I had a computer program) merely discovered what was already known about H2 (because of the abundant genetic information available for mice).
H2 is a two-locus system, exactly analogous, at the foundational level, to the two-locus HLA model discovered later by the computer program.
The page you are now reading was added to my web site in 2018. The reason was to clarify my position as a sort of whistle blower in relation to the biomedical research establishment, and to support the need to heed the pioneers of the science behind the connection of nutrition and disease. The establishment has historically ignored any connection of diet with disease, and only recently accepted a small connection – not enough to revolutionize the "standard of care".
Yes, the establishment's intransigence may be due to the difficulty people in general have with the unexpected and unfamiliar and the investments specialists have with their current working assumptions; looking back with a thirty-year perspective, I am no longer surprised that no one wanted to do an experiment that would jeopardise their own conclusions. But the establishment has also been corrupted by the pursuit of money, prestige, career – from companies that want to sell you something, to the scientists doing the research. I recently got an email from an organization (they're called "head hunters") that asked if I wanted to recommend someone for the position of Dean of the College of Sciences at a major university. They supported the quality of this college by stating how many millions of dollars the faculty brought in from outside sources. That's where university administrators are, all too often, coming from. A researcher that finds and publishes a result that his or her funding company doesn't like will not get funded again. And what then will happen to the researcher's career?
In the old days, before huge amounts of government money were available for research, scientists did research because they were interested in people’s health, or because they were interested in advancing our knowledge. Some led lives of paupers, putting all their resources toward their work. Now I found an abundance of people primarily interested in money – the key to academic advancement, prestige, career.
After letting my own biomedical research go, I turned to teaching students (who were training to teach mathematics) about deductive mathematics. (See the introduction to the online text in the "mathematical proof" section of this site.)
There was one of my papers in biomedical research that had received a lot of interest, shown by requests from around the world for article reprints. I stopped counting at 800. The last, lonely reprint request that I ever had, however – and oddly enough – was for a paper on the action of the hyperoctahedral group.