[ Question ]

Anyone have experience with metabolic network modelling that would be willing to help me out?

by CraigMichael1 min read14th Jun 20212 comments

3

World Modeling
Frontpage

In 2019 I had Megavitamin-B6 syndrome. I'm curious about the most common situation where people get it (chronic supplementation vs. acute overdose), and I have a hypothesis about how that may come about. Ideally, I'd like to see if it pans out in a model, and if it does maybe see if I can get some attention from interested biologists. But my background is in CS, not biology and I feel like I'd have a lot to learn. So anyone who would collaborate or just get me pointed in the right direction would be helpful. 

In short it seems the chronic supplementation does factor in, but it also seems that plenty of people supplement for a very long time (more than 2 years) and never have an issue, sometimes at or above NOAEL or the even the LOAEL (100mg/day and 200mg/day in the US). 

From many of the people I've talked to (and my own experience) it seems like something happens where someone takes a consistent dose of supplemental b6 for a year or more (at a reasonably high dose, say, 25mg/day or more) then at some point they take something else with b6 (or maybe are taking a poorly calibrated supplement that has more b6 than advertised) and then they start to develop toxicity, even if it the total dose is still below the "NOAEL."

So, my guess is that metabolic pathways adjust to receiving some daily dose of b6 and then become inflexible somehow (like a homeostatic thing) and can't ramp up the bandwidth/rate to process it in to 4-pyridoxic acid, the form where it's excreted. 

Soo... I kind of want to see if that's plausible in a realistic model. 

New Answer
Ask Related Question
New Comment

1 Answers

I haven't looked into vitamin B6 in before and just looked at the Wikipedia article today. I start with general things:

A lot of what happens with metabolic pathways well modeled with control theory. 

The body has an "idea" of how much it needs of a given substance and then up and downregulates genes to get to the targeted amount. 

It's possible that the presence of a vitamin activates a gene that then produces an enzyme that process the vitamin.

It's possible that the presence of the vitamin blocks a gene that produces an enzyme that produces the vitamin or helps making the vitamin available from food.

In addition to regulation like that where the effects are immediate, methylation also allows the body to switch off genes for longer amounts of time. 

To look at Vitamin B6 in particular you get https://www.genome.jp/kegg-bin/show_pathway?map00750 as a map of the related process and genes. It's possible that some of them are not expressed because they get methylized when they aren't needed.

If you want to go deeper into the topic, learn what the things in the pathway happen to be and what we already know about how the related genes are regulated. You can search through existing papers for everything involved. 

If some of the involved parts of the pathway don't have Wikipedia pages, the task of writing a Wikipedia page about them to explain what information there is about them is a useful way to study them. 

There are a lot of different genes and there's more reserach interest in some of them then others. If you can summarize on Wikipedia what a gene does for which there currently isn't that much research interest you might produce an article that's useful for anyone who cares in the future about knowing what the gene is about.

 The KEGG reference here is good. Thank you! I'm hoping to get to a point where I could take something like what's in KEGG and kind of play with it? Like if you've used something like Emergent to model the brain, something like that but with metabolic pathways. 

Like take a set of metabolic pathways, play with amounts of chemicals, enzymes, genes that are and aren't expressed, etc and use it run a simulation and test hypotheses? Would like to see what ways I could get pyridoxine, pyridoxal, pyridoxamine, etc to "build up." 

Like take all the e... (read more)