Microbiome Suggestions for Autism — Cross Validation

Microbiome Prescription uses over a million rules to generate suggestions on improving the microbiome and hopefully reduce or moderate autism behavior. All of the sources of the rules are studies on the US National Library of Medicine. Microbiome Prescription also can provide the complete evidence trail for every suggestion! That is, where — precisely– is all of the information coming from — none of it is private personal opinion or speculation.

Cross Validation

Cross validation is the process of taking one set of information to generate forecasts or suggestions and then look at a totally independent source of information to see if the forecasts and suggestions are valid, reasonable, and appear to help individuals with the condition. I have done that for several conditions with very good results. NOTE: Everything is generated by code — code that I prefer to improve or correct. I have no personal stake in the suggestion, nothing to defend.

Source of Suggestions

Simple, šŸ„£ Candidates which is based on bacteria shifts reported from studies for autism: šŸ¦  Taxons. These two links are on the Medical Conditions with Microbiome Shifts from US National Library of Medicine page

The process is simple: some items may have been tests in trials for autism, some have not. If it has been tried, we see what the result in and provide a link to the study (open data!! no “trust me, I am an expert” hype)


  • Five items with no information
  • Two items with weak information (not PubMed)
  • Twelve items with confirmed information (PubMed)
  • One item that is complex/questionable – depends on the child’s DNA

The goal / objective of Microbiome Prescription is to make suggestions that are more likely to help than to hurt. That goal seems to be accompanied. A secondary goal is to suggest items that have not been studied but modelling suggests that it may be of benefit to try. We have 5 such items above.

Remember these are GENERIC Suggestions for GENERIC Autism

The results of an individuals microbiome will be different — there are many variants and subsets for Autism. Each variant will tend towards their own set of variations for the microbiome. Using an individual’s microbiome sample will get suggestions unique for them.

Postscript ā€“ and Reminder

I am not a licensed medical professional and there are strict laws where I live about ā€œappearing to practice medicineā€.  I am safe when it is ā€œacademic modelsā€ and I keep to the language of science, especially statistics. I am not safe when the explanations have possible overtones of advising a patient instead of presenting data to be evaluated by a medical professional before implementing.

I cannot tell people what they should take or not take. I can inform people items that have better odds of improving their microbiome as a results on numeric calculations. I am a trained experienced statistician with appropriate degrees and professional memberships. All suggestions should be reviewed by your medical professional before starting.

The answers above describe my logic and thinking and is not intended to give advice to this person or any one. Always review with your knowledgeable medical professional.

Leave a Reply