⚠️ Under construction.
LifeNome Platform
Platform Data Model

Intro

The Data Model section outlines the structure and relationships of key entities within the LifeNome Core Platform. This framework is crucial for processing and analyzing both genetic and non-genetic data, enabling the delivery of personalized health and well-being insights.

Sample

In the LifeNome Core Platform, a “Sample” refers to an anonymized abstraction of an individual from whom genetic and non-genetic information is gathered. Subsequent analysis is performed to provide a holistic view of the individual's health and well-being, allowing for a comprehensive evaluation that extends beyond genetics to include lifestyle, environmental, and health-related factors. Each Sample is represented by a data object with an abstract (anonymous) identifier, ensuring that no personal information about the individual is included.

Sample Attributes

Sample Attributes encompass a diverse array of attributes and data points linked to each Sample within the context of the LifeNome Core Platform. These attributes may include genotype, as well as demographic, lifestyle, or wearable data, collectively offering a comprehensive overview of each Sample's identity and characteristics.

Genetic attributes, often referred to as genetic variations, are stored in a Genotype data object that is linked to a Sample. This data object holds information about the individual's genetic makeup.

Non-genetic attributes are optional data points used to store additional information about the Sample. Unlike genetic attributes, non-genetic attributes are not predefined by the LifeNome Core Platform. Instead, they can be configured at the Tenant level. For example, one tenant may choose to collect information about a Sample's gender, age, weight, and height, while another tenant may opt to gather data about dietary habits. This flexibility allows customization based on the specific needs and preferences of each tenant using the platform.

Traits

An individual's traits, encompassing both genetic and non-genetic factors, collectively define the distinctive characteristics and qualities that contribute to their identity and physical or psychological makeup. Genetic (or DNA) traits are inherent traits passed down from biological parents, shaped by the individual's genetic composition and encoded within their DNA. In contrast, non-genetic traits arise from external influences, life experiences, and factors unrelated to an individual's genes.

A Trait object represents a quantifiable characteristic of an individual. This quantification process relies on a trait-specific computational model utilized by Assessment Engine. The result of this computational process is referred to as a Trait Assessment.

The inputs provided to the Assessment Engine include values derived from the individual's attributes (genetic and/or non-genetic). Additionally, values from other Trait Assessments may contribute to this comprehensive evaluation. This approach ensures a thorough understanding of various traits within the LifeNome Core Platform.

DNA Traits Assessments

DNA Trait Assessments are quantitative indicators that offer insights into the likelihood or susceptibility of individuals to particular traits or predispositions, depending on their genetic attributes. These measures are generated through an analysis of an individual's genetic composition. They function as predictive indicators, signaling the probability or risk associated with various traits.

A Trait Assessment data object stores the calculated Trait Assessment results for a particular Sample. This data object serves as a repository for the numerical assessments, providing valuable information about an individual's trait characteristics and potential predispositions. It includes essential details such as the trait code, name, description, connotation, type, categories, assessment specifics, coverage information, recommendations, and any additional properties associated with the trait assessment, all organized in a structured format as follows:

  • trait_code (string): The code identifying the trait.
  • trait (object): Information about the trait.
    • code (string): The code identifying the trait.

    • name (string): The name of the trait.

    • description (string): A short description of the trait.

    • connotation (string): The connotation of the trait, which can be one of the following:

      • "good" - Indicates a positive trait.
      • "risky" - Indicates a risky trait.
    • assessment (object): Information about the trait assessment.

      • score (number): The assessment score as raw score computed using LifeNome's algorithms, serving as the basis for determining the assessment level.
      • level (integer): Maps the score to an assessment level (0, 1, 2), which can be interpreted as follows:
        • Level 0 (low predisposition): Indicates that individuals are unlikely to have a predisposition for this trait.
        • Level 1 (moderate predisposition): Suggests that individuals have a somewhat higher than average predisposition likelihood for this trait.
        • Level 2 (high predisposition): Indicates that individuals have a significantly higher likelihood of this trait compared to the average person in the reference population.
      • percentile (number): The percentile score for the trait indicates where an individual's genetic predisposition likelihood for the trait stands compared to the reference population.
    • coverage (object): Information about predisposing genetic variants (SNPs) relevant to the trait assessment.

      • lookup_snps (integer): The total number of genetic variants important for the trait.
      • snps_found (integer): The total number of genetic variants important for the trait found in an individual's DNA.
      • inhibiting_snps (integer): The number of genetic variants that inhibit an individual's predisposition likelihood for the trait.
      • contributing_snps (integer): The number of genetic variants that contribute to an individual's predisposition likelihood for the trait.
    • recommendations (array of objects): An array of recommendations related to the trait assessment level.

      • type (string): The type of recommendation.
      • title (string): The title of the recommendation.
      • content (string): The content or details of the recommendation.
    • axengine (object): Additional properties related to the trait assessment, identifying computational engines used for the assessment score computation.

Genotype

A Genotype refers to the genetic composition of an individual, specifically the unique combination of genes found in their DNA. It encompasses the distinct set of genetic information that individuals inherit from their biological parents, and this genetic makeup can profoundly influence their physical characteristics, susceptibility to diseases, and various other traits.

Obtaining an individual's genotype can be achieved through genetic testing, often referred to as genotype analysis. This process involves examining an individual's DNA to identify specific genetic variants, mutations, or alleles associated with particular traits or conditions. Alternatively, genotype information can be obtained through methods like whole-genome or exome sequencing, which capture an individual's complete DNA sequence, including all genes and non-coding regions within their genome.

In the context of the LifeNome Core Platform, the Genotype is considered a genetic attribute of the Sample, serving as a crucial component in understanding an individual's genetic characteristics.

The Genotype is represented as a data object that establishes a link between the Sample object and raw genotype data files. These raw genotype files are securely stored in a Platform Blob Storage (PBS). There are various ways to import these raw genotype data files:

  • Direct import of files to PBS,
  • HTTPS file upload initiated via Core API,
  • Files resulting from DNA kit processing (integration with lab analysis provider).

These raw genotype data files serve as essential inputs for computing DNA Trait Assessments, enabling a deeper understanding of an individual's genetic predispositions and characteristics.

DNA kits

A DNA kit is a comprehensive package or collection of components and materials designed to simplify the process of collecting, preserving, and transporting biological samples that contain DNA. These kits streamline the acquisition of high-quality DNA for subsequent genetic analysis and genotyping.

Within the LifeNome Core Platform, a DNA kit management capability efficiently oversees the inventory of DNA kits. This capability also facilitates seamless integration with external laboratories, such as Labcorp or Gene by Gene, which offer genotyping services.

The DNAKit object represents the actual DNA swab used for collecting DNA samples. These DNA kits are provided by LifeNome and distributed to Tenants, who, in turn, supply them to their end-users. Once a DNA swab is used, it is sent to a processing lab. This laboratory conducts DNA analysis and generates raw genotype data files. These data files are a fundamental resource for further genetic assessments and analyses conducted by LifeNome.

Reports

Reports in the context of the LifeNome Core Platform are structured documents designed to present trait assessment-related computation results in easily readable formats, including PDF or HTML. Reports integrate the output of various computations with relevant descriptive and graphical content, all while adhering to predefined layout templates.

The Report object includes a unique report identifier, details about associated documents (which can be in PDF or HTML format), the document's language, and the timestamp when the report was created.

Recommenders and Ranked Items

Recommenders within the LifeNome Core Platform generate personalized rankings of various rankable items. These items can include a wide range of products, services, or content, such as foods, beauty products, exercises, and more. The recommender system uses algorithms to suggest or rank items that are likely to be of interest or relevance to individuals based on their preferences, behaviors, or traits.

Ranked Items are part of the Item Collections; for example, a skincare products collection could be used during the deployment of a skincare recommender subsystem.

Each Ranked Item in the Recommender output has:

  • id which uniquely identifies that particular item,
  • full data whose scheme might differ for different item collections,
  • ranking score with subscores.

The ranking score is the total sum of all subscores, while the subscores show how individual features of the Ranked Item impact its ranking. The ranking score is not normalized and can be used for a general overview of how suitable a particular Ranked Item is for the given sample, relative to other items in the same Item Collection. Subscores can be used to investigate which particular features are most important.

Note that the Ranked Item data schema might differ across various item collection types. For example, skincare products have brand attributes, while items in the exercises collection have instructions attributes.

Questionnaires

The Questionnaire model within the LifeNome Core Platform offers a flexible and efficient way to collect user inputs and compute various assessments based on those inputs. It supports the collection of non-genetic attributes, which can be critical for personalized evaluations.

The platform allows tenants to list all configured Self-Assessment Questionnaires (SAQs), retrieve specific questionnaires by code, submit user responses for analysis, and access the necessary JSON schema for rendering forms. Additionally, the platform provides tools to fetch the OpenAPI schema for SAQ microservices and the schema that defines valid SAQ answers, ensuring that all data conforms to expected formats.

Assessment Services

The LifeNome Core Platform's Assessment Services model enables the execution of various assessment computations that combine genetic and non-genetic data to generate insights. These microservices expose a range of assessment computation capabilities, allowing for the integration of personalized health and wellness analyses. An example is the Longevity Assessment Service that calculates longevity scores based on lifestyle, medical history, and dietary/fitness habits.

Tenants can list all available assessment services, retrieve specific service details using a unique code, and run assessments by invoking the appropriate service with the relevant data. The results of these assessments are indices used to provide tailored recommendations and insights based on the individual's unique genetic and non-genetic profile. Additionally, the platform provides access to the OpenAPI schema for each assessment service, facilitating seamless integration and customization within different applications.