EVOgnostic, Evotec’s diagnostics division, stratifies patient populations in health-to-disease maps and discovers most informative biomarker panels based on comprehensive clinical and molecular patient data to drive the drug development process, and ultimately empower clinical decision making and companion diagnostics. 

EVOgnostic partners on and provides unique molecular biomarkers in complex diseases and defines specific patient subgroups in these multifactorial diseases that can unambiguously correlate with biological events to validate drug targets, diagnose disease, monitor disease progression, design clinical trials, and predict drug response and clinical outcomes.

EV Ognostic PC graphic 02 230825

We combine all of PanOmics applications including transcriptomics, deep proteomics and metabolomics in human sample cohorts, animal models and clinical trial samples at cell population and single cell level with machine learning, to accurately characterize patient populations and discover disease specific signatures. Acquisition and computational analysis of multi-omics data from a combination of tissue and different body fluids allow us to obtain a holistic picture of the underlying drivers of disease for each individual patient. 

The discovery and use of our specific objective biomarkers and multi-omics disease insights are fundamental from the early stages to the late stages of the drug development process to reach desired clinical trial endpoints. These biomarkers will inform regulatory and therapeutic decision-making regarding candidate drugs and their indications in order to bring new medicines to the right patients faster.

EVOgnostic Biomarker Discovery Workflow

EV Ognostic PC graphic 230825
  1. Heterogeneous Patient Population
    Numerous complex diseases with poorly understood aetiologies are characterized by diverse causes that result in similar clinical presentations. Moreover, patients exhibit variations in disease progression and symptom profiles. There is a critical demand for specific biomarkers that can effectively identify individuals at risk of developing a disease, predict treatment response, and assess disease progression.
  2. Comprehensive Molecular Patient Database
    As a fundamental resource for achieving a comprehensive understanding of diseases, our Molecular Patient Database integrates clinical and multi-omics data. This extensive database comprises information from over ten thousand patients, including existing data, ongoing patient cohort collections, and publicly available data. These cohorts encompass both cross-sectional and longitudinal studies.
  3. Machine Learning
    Traditional data analysis methods often struggle to discover biomarker signatures within the complexities of omics data. To address this challenge, we leverage machine learning as a pivotal technology to unveil the most suitable biomarker signatures. Our approach involves the comparison of various algorithms, enabling us to identify the algorithms that are best suited for the task.
  4. Omics-based Patient Stratification
    By utilizing machine learning and statistical methods to identify disease-specific signatures, we enable the omics-based stratification of patients. Our objective is not only to distinguish between healthy and diseased individuals but also to assign disease scores that position patients along a continuum from health to disease. We refer to this approach as the Health-to-Disease map (HDM). This continuum extends beyond disease onset, providing insights into disease risk scoring, relapse prediction, and identifying additional health states.
  5. Feature Reduction
    Initially, machine learning models utilize complex PanOmics data. To narrow this down to most informative disease signatures and corresponding biomarker panels, we employ feature reduction techniques, ensuring the selection of a concise panel of biomarkers suitable for clinical applications. We strive to maintain high diagnostic performance while minimizing the number of features in the biomarker signature.
  6. Biomarker Identification
    Upon receiving a list of top candidate biomarkers from our machine learning analyses, we apply our disease expertise to evaluate their performance as a biomarker panel. This candidate list provides the basis for developing laboratory-developed tests (LDTs), in vitro diagnostics (IVDs), or other diagnostic products and services intended for clinical decision support.
Christiane Honisch

Dr Christiane Honisch

SVP Head of Diagnostics

Logo Evotec white
Evotec has the right technologies & disease understanding to meet our partners' evolving needs: a comprehensive disease knowledge at the molecular level, cutting-edge technologies & platforms to translate this expertise into effective precision medicines.