Abstracting the complexity

How can we make sure we do not get seduced by the pitfalls of other companies that try to use one approach and fail to gain enlightening knowledge by that way

Now, let's dive into the specifics of how we gamified the microbiome research approach at NostraBiome. Our goal was to simplify the complexity of the microbiome and provide an engaging and interactive way for researchers, including non-medical team members, to explore and understand the microbial world. We recognized that abstracting important elements visually was key to ensuring that everyone on the team could understand the set and contribute to the research process. Through a combination of interactive visualizations, machine learning algorithms, and curated databases, we aimed to create an immersive experience that enabled the whole team to gain a deeper understanding of the microbiome and develop new insights that drive progress in the field.

We needed a simple way - so that a kid will understand how the microbiome and gut equilibrium works and will be able to visualize.

! PLEASE READ THE BELOW AS CAREFULLY AS POSSIBLE AS IT IS FUNDAMENTAL FOR OUR APPROACH TO MODULATING THE MICROBIOME AND BEING SUCCESSFUL SO FAR WITH VERY LITTLE DATA

What if we say that the GUT is like the environment in a game

with a set of rules enforced by some rulers( rules.gut) & (game-rulers.gut)

a set of active elements or players called bacteria of different types ( bacteria.gut)

who do different activities (activities.bacteria)

and who all produce certain signals through their activity (signals.bacteria) aka metabolites .

Let's assume each signal can be made in a multitude of colors by each player ( colors.signal)

those signals can be helpful for the environment of the game, but can also be harmful depending on the quantity and moment they are released by each player. (good-signals & bad-signals)

those signals can interact with other players (triggers.signal) telling them to do something,

those signals can also make other players send certain color signals to others and change their activity in the game.

The importance of the game is that the setup of signal colors happening in the game environment has a certain configuration or multiple configurations that are correct and tolerated by the rulers. If the signals are bad the ruler will try to correct the player.

If the signals are not having the correct configuration or are tolerated by the environment, the game rulers will try to kick out some players or put them on hold to make sure he continuously gets the correct sets of color signals.

although

the players can hack the game, and send their signal colors in a way that the game doesn't know who to kick out anymore

and

because the game is hacked and the signals of certain players are affecting the environment, the game itself will end up trying to do different things to eliminate the players without being right. This will eventually lead to a mess that will affect the game itself, the other honest players, and the environment where the game is played.

As game supervisors, we can interact with the game and help the game re-establish its activities when it is hacked by understanding who are the bad players, what bad activities they do, what bad signals and what good signals they produce, and how those signals affect the game balance and the environment.

Also

by gradually shutting down the bad players we give the chance to the game environment to establish balance back and let the rules of the game prevail.

How do we help the game regain access to its rules and not harm the game environment?

This was our starting thinking point and framework of attacking such a dynamic problem/ setup

Of course, we did not build a game, but we used this simple visualization of elements to determine which our next steps should be and how we architect the system in a way that is efficient with small amounts of data, is based on resilient research, has impactful results and ultimately will learn from more users and drive shorter loops of understanding.

Therefore we have this high-level system schema

Now the pre-requisite challenges of this paradigm are the following:

1. There are so many different types of players

2. There are so many game rules

3. There are so many different signals to be monitored with a huge range of color nuances for each player

4. There are so many rulers behaving differently. Sometimes some of them also misbehave and get tricked by certain signals.

5. Can we determine the real-time monitoring of this system with today's existing tools and technology? NO

The above being mentioned makes it rather impossible to build a foundation, yet there are medical facts we can rely on and extrapolate knowledge to reverse engineer and generate more knowledge and actionable insights.

In this case, how can we navigate through this tremendous set of permutations? What can we do?

1. Focus on what we know from science about each bacteria

2. Focus on the combination of player-bad-signal-ruler-action and how it manifests in the system

3. Introduce a new concept ( signal_damage_result) that basically refers to what a bacteria doing certain signals of a certain type will produce to the environment of the host. This will be translated as Sympthoms.

4. Identify and quantify the total bacteria and number of each ( Megategenomics Shutgun,16 s DNA and qPCR) to understand the statistical distribution between each player and the signal_damage_result on a host measurable via blood tests, inflammation markers, and user-declared symptoms.

5. Instead of trying to monitor the system in real-time try to look at the results of the game hack effects, combine them with knowing who are the players, and try to reverse engineer a set of rules for eliminating the bad players who lead to that signal_damage_result

6. We also know that IBD is clearly a manifestation of a system being hacked for too long where the rules, rulers, and players in the system are totally messed up and the signaling perception makes no longer sense.

Therefore the prerequisites for having a functional setup that can at least eliminate at first the players that violate the rules is to have a detailed list of all the bacteria identified in human stool samples and create backed medical research on individual bacteria findings database that will serve as foundation element to build a predictive algorithm to modulate the microbiome with the purpose to understand the bad players think how they can be eliminated and later evolve into going more granular in understanding the signaling color palette of each bacteria as soon as enough labeled data will be available.

The database will be an evolving IP element of the company and will be strengthened day by day by the findings of our technology, but will also be manually annotated by the findings of other researchers.

As the purpose of this document is to explain further the approach of the algorithm please do not expect that we will disclose in the next pages the exact structure and very detailed steps but we present more of the high-level architecture design behind the algorithm, its relation to science-based evidence knowledge, usage of science data and it expands further the presentation to the already introduced paradigm of the gut being like a game environment.

IBD is nothing else than the result of a hacked game where different players manifest a totally unique set of combinations for each host.

The goal of the NostraBiome algorithm is to do 3 things as macro responsabilities

1. Identify the bad players creating bad signals

2. Define what is the most appropiate treatment to eliminate all the bad players

3. Ensure that the host has enough time to start getting back to normal the rule-rules-signals mechanisms, minimal damage to good guys so the game can be played.