Genialis Supermodel

Genialis Supermodel is a large molecular model (LMM) of cancer biology, trained on hundreds of thousands of RNA sequencing samples drawn from preclinical studies, single-cell data, and globally diverse patient records. Designed for translational and clinical research teams in oncology, it maps each new patient tumor sample to a comprehensive biological landscape, revealing underlying drivers and vulnerabilities that inform treatment decisions and biomarker development.

At its core, the Genialis Supermodel delivers therapeutic intelligence by interpreting the molecular biology of tumor samples. Its outputs are biomodule scores — algorithmic, transcriptomic representations of diverse biological phenomena, states, and processes. Technically, the Supermodel is a compendium of phenotypic embeddings that serve as input features for machine-learning modelling of predictors, enabling teams to interrogate virtually any prediction task related to response, mechanism, durability, indication, or line of therapy.

What Biomodule Scores Enable

• Distinguish responders from non-responders to a given therapy
• Generate hypotheses about mechanisms of response and resistance
• Suggest likely combination treatment approaches
• Uncover novel targets for cancer therapy

Four-Phase Workflow

1. Primary analysis — processes raw RNA sequencing data and produces gene expression profiles
2. Data harmonization — covers data quality assurance, pre-processing, batch effect detection and mitigation, and integration with proprietary and third-party licensed data
3. Biomodule scores — transcriptomic embeddings computed with the Genialis Supermodel that transform harmonized sequencing data into a low-dimensional biological space
4. AI predictions — made by applying existing predictors or training new AI models on biomodule scores to generate clinically relevant insights

Key Software Components

• Genialis Expressions — a robust, scalable cloud software that captures analysis metadata and handles primary processing of sequencing data through a suite of validated pipelines
• Genialis Precision Medicine SDK — prepares data for machine learning using data normalization, a system of preprocessors, and a framework for batch effect detection and removal
• Genialis Supermodel — transforms harmonized sequencing data into a low-dimensional biological space using biomodules
• Predictors — AI models developed to predict response, prognosis, or other clinically relevant outcomes; these can be trained by Supermodel licensees or contracted and licensed directly from Genialis

Applied Biomarker Examples

• Genialis krasID — the first and only biomarker capable of accurately stratifying KRAS patients by clinical response
• DDR biomarker — an algorithm system for DNA Damage Response agents designed to identify responders to WEE1 inhibition

The Genialis Supermodel can be accessed on Genialis-managed cloud infrastructure or deployed within a customer's own environment, with rich APIs enabling bundled or independent integration of each software component into existing research and AI ecosystems. Genialis also provides AI-enabled expert services to support technology integration, data processing, and AI design, modelling, and interpretation.