MONADIC DNA

OPEN, SECURE GENOMICS FOR ALL

Introduction

Genetic data is becoming increasingly important in healthcare with CAR-T cell therapy, CRISPR and personalized medicine already enjoying early successes. The costs of genetic sequencing are declining at a Moore's law rate^[1]. Newborns across the world are being genetically tested by law^[2]. The stage is set for pervasive sequencing for public health and medical care. Yet, not only is there a paucity of genetic data available for research purposes^[3] but the existing personal genomics services with the largest datasets have failed to open and monetize their data^[4].

The Monadic DNA project, consisting of two apps for bootstrapping an ecosystem and an open, decentralized protocol for sustaining it, seeks to revolutionize the genomics space by immediately unlocking the latent value of genomic data for personal genomics enthusiasts, patients, study participants, academic researchers, pharmaceutical companies, citizen science projects and biohackers. By following a bottom up approach with strong, well-placed incentives and frontier privacy-enhancing technologies such as FHE and MPC, we aim to make millions of new genetic sequences available for analysis and prepare the ground for the secure storage and processing of billions in the future. This will unlock a range of transformative applications in medicine, lifestyle optimization, healthspan extension and unforeseen disruptive applications that emerge from the open and decentralized nature of the platform.

Exemplar Applications

Personal genomics has immediate and practical utility for people so there is no solution looking for a problem. We therefore start with useful applications to deliver value to users with a strong product focus and then back out the technology and underlying protocol accordingly.

Monadic DNA Mobile App (Code Name: Monadic Vault)

Prototype of the Monadic DNA Mobile App

Monadic Vault is meant for regular users who have been underserved by incumbents such as 23andMe. The app allows users to either bring their existing data from other genomics services or get freshly sequenced by Monadic DNA, store their data securely, get health, ancestry and lifestyle insights and monetize their data.

The app uses fully homomorphic encryption (FHE) to encrypt the user's data and uploads it to our servers for now and, eventually, to a decentralized store such as nilDB. Using FHE, we can run computations on the server side without ever needing to decrypt the data. The results are sent back to the users' app where only the user is able to look at or share the results.

As 23andMe with its 16 million users has a doubtful future^[5], both our marketing and onboarding experience will focus on acquiring these users with strong messaging and hand holding through the data migration process. Users may be given five free insights upon onboarding and could then receive a new insight through a mobile notification every two weeks to maintain engagement. A marketplace of insights will also allow third parties to bring in and monetize their expertise. Finally, other apps, especially in health and fitness, will be able to connect their accounts to their users' Monadic DNA account to combine DNA insights with other datasets. For example, Whoop could recommend either sprinting or jogging to their users depending on their genetic profiles.

Monadic Pools ^[6]

Monadic Pools is targeted towards more savvy users who have experience with DeFi apps, NFT marketplaces and memecoin platforms. On this platform, people such as researchers and biohackers create funded research pools into which users deposit their genomes in return for token rewards. The app emphasizes virality and engagement. For gamification, participants earn points for various activities on incentivized leaderboards.

The primary participants are:

• Pool creators who want to run a study or analysis
• Users who want to monetize their genomic data through pool rewards
• Funders such as disease communities, biohacking groups or social media groups who want to fund pools
• Delegators who act on users' behalf to enroll them into pools

Rough sketch of the Monadic Pools Application - Incentivized Research pools

Pool creators specify the name of a pool, along with a funding amount, required attestations (for gating), premium attestations (for extra rewards), number of genomes needed, the code to be run and the visibility of the results. The creator can wholly fund the pool themselves or appeal to the community to fund it. The latter might be useful for, say, the active community of patients of the rare disease Ehlers-Danlos syndrome (EDS) where a researcher creates the pool and asks on the community forum for funding. Required attestations can be used to gate the pool to attested participants from a specific region or those with specific traits or NFTs. Premium attestations multiply rewards for holders of specified traits or NFTs so that an EDS pool can, for example, incentivize sufficient participation by actual EDS patients. The research pool's visibility controls whether the results of the research pool are public, gated or private. When users sign up, they may be required to fill in a questionnaire to provide metadata for research purposes. Once a pool is fully funded and the required number of genomes is locked in, the code is executed, the results are attested and tokens are paid out to the pool creator, participants and the token holders of the underlying protocol.

This app will provisionally use various technologies provided by the Nillion project. nilDB, a decentralized encrypted database using multiparty compute (MPC), will be used to store and secure the data. nilQL, a query language, will be used to securely extract the required subsets of users' genomic data which is then deposited into trusted execution environments (TEE), where the code for pools will be run. nilAI, which integrates nilDB and TEEs out of the box, will be used for research pools which need to use AI foundation models.

To prevent drop offs in engagement and reduce friction, we are introducing the concept of genomic delegation. While the default path is for the user to maintain complete confidentiality and control over their data, we offer those users who trust us (or any other entity within the ecosystem) the ability to delegate their data, i.e., upload a copy of their data under the control of an account controlled by the delegate. This lets the delegate find suitable pools for users and auto enroll them, thus ensuring constant rewards for users without their constant active participation. In return, the delegates earn a fraction of the rewards as a fee. Delegates compete for users based on their reputations, costs and missions.

Once Monadic Pools is fully implemented, the marketplace and monetization features of Monadic Vault simply become a curated frontend for Monadic Pools functionality, especially delegation.

Data Flow

Ikai DAO is a gatekeeper which accrues value from genomic storage and compute

Once a user provides a saliva sample and sends it in by mail, a verified lab sequences their DNA and posts it to the API of a verified app, e.g. Monadic Vault. Most labs allow only anonymous identifiers to be provided and convey the sequenced data to the apps accordingly. Thus, while the data is not encrypted at source, it is private and anonymous. In the future, Monadic DNA may operate its own labs to provide encryption from source.

Apps allow users to claim their data using a predetermined claim code, upon which the data is downloaded to the user's device to be encrypted and stored on a nilDB cluster. Such clusters, built using Nillion blind computing technology, allow data to be stored encrypted over multiple database nodes using MPC technology so the security and integrity of the data is maintained even if only a single node on the cluster is honest. As a risk mitigation measure, Monadic Vault will retain a compressed and encrypted copy of the DNA data within app storage as a fallback in case the data needs to be migrated to later versions of the technology.

The Ikai DAO controls which queries can be run on the cluster and by whom. Thus, any app which wants to operate on users' secure genetic data will have to make a proposal to the DAO to be attested and, further, have their template queries for fetching data attested as well. This ensures only the minimal set of required data is shared and only authorized people have access after paying the protocol. nilDB clusters are queried using nilQL and access is controlled using industry standard JSON web tokens (JWTs), which the Ikai DAO will incorporate into its governance process using onchain attestations.

Credentialed apps such as Monadic Pools can query out the subsets of users' data that they need for processing in pools or gating access to apps, etc. For example, a research pool studying athletic performance may only need a few genetic markers. Before querying, however, apps have to reserve a trusted execution environment (TEE) which allows code to be run on data inside a secure enclave in the cloud. Once an app reserves a TEE, it gets a JWT to query out the users' data as authorized. After receiving all the data it needs for any activity, e.g. a research pool on Monadic Pools, the app executes the code inside the TEE and fetches the result. If the result is a genetic trait for a user, then the app can write it into the nilDB cluster and create a corresponding onchain attestation to validate the stored trait for other actors of the protocol.

A similar mechanism holds for delegation. If a user chooses an authorized delegate, their app records this onchain which enables the DAO to give carte blanche permissions for the users' nilDB data and traits to the delegate who can then use it for ecosystem actions, e.g., auto-enrolling the user in research pools on Monadic Pools.

Nillion's upcoming nilAI product will combine nilDB and TEEs to offer scalable and secure AI processing of encrypted data. However, if needed for better suiting our needs, we can build a bespoke solution using FHE for secure storage and computation.

There is an alternate data flow using fully homomorphic encryption and decentralized storage for the Monadic Vault rollout but, for the sake of simplicity, this information is being omitted. Note that the Monadic Vault app will be using the alternate flow until the Ikai protocol is ready.

General Prospects

App Ecosystem

While Monadic Vault and Monadic Pools will bootstrap the ecosystem and demonstrate the latent value of genomic data, we expect fresh, unprecedented apps by developers, biohackers, academics and major companies to grow the ecosystem in unexpected ways. As an indication, these might include:

• Apps that combine genetic insights with other biomarkers to provide holistic advice
• Dating apps which match people based on compatible fun traits
• Video games in which a gamer's character is based on their traits
• Music generation based on people's genetic sequences
• AI agents that use genetic traits and other health data to offer custom insights to users

Enterprise

As mentioned above, newborn children in all fifty US states are now required by law to be genetically tested. This is also true in many other industrialized nations and will soon also be true in populous countries such as India and China as sequencing costs fall. Not only are millions of children now being sequenced but this data is required to be held by the state well into these children's adulthoods. It is only a matter of time before state-run data stores are breached amidst public uproar. The end-to-end encryption flow pioneered by Monadic DNA will then be adopted to minimize these breaches and keep the data secure while still facilitating public health research.

Likewise, medical systems, pharmaceutical companies, IVF clinics, insurance companies and genomics providers will have increasingly large exposure to the sensitivity of genetic data, especially as it becomes a standard part of healthcare and therapeutics through CAR-T, CRISPR, personalized medicine and other pioneering technologies. While individuals are not significantly privacy-conscious, at the societal level there are large penalties for breaches and great reputational loss. Just last year, the CEO of United Healthcare was brought before Congress^[7] to answer for a major breach. 23andMe was also hacked recently^[8], leading to great user unease and reluctance by many to use their services. Once again, our protocols, technology and expertise will facilitate the storage, sharing and processing of genetic data within and between enterprise players.

Conclusion

For far too long, genomic data has been jealously guarded from users by floundering companies with no security or privacy guardrails. Monadic DNA seeks to create a data governance DAO and enhance it with cutting-edge encryption to create an open protocol which can unlock the latent potential of sensitive genomic data. We aim to facilitate an app ecosystem consisting of two-sided marketplaces, user-friendly personal genomics apps and many other novel applications we can't yet conceive of. The success of the protocol can come from the general utility of the ecosystem, the success of a single successful application or the development of novel IP which enables outsized returns from billions getting sequenced over the coming decade.