Skip to content

Introduction

Horizon 2020 ALTERNATIVE Project

Horizon 2020 research project ALTERNATIVE develops an innovative platform and an integrated approach for testing and assessment of the potential of chemicals to induce cardiotoxicity. The novel approach will strengthen the capacity of regulators and industry to prevent cardiotoxic co-exposures to industrial chemicals and pharmaceuticals in an effective way.

Purpose of the Cloud Data Platform

The Cloud Data Platform is the cornerstone of the ALTERNATIVE project, designed to facilitate the project's core objectives. It serves multiple purposes:

  1. Central Hub for Data and ML Model Exchange: The platform acts as a central repository, enabling efficient and seamless exchange of data and ML models among consortium partners. This ensures that all participants have access to the latest resources, fostering collaboration and accelerating research progress.

  2. Innovative Environment for ML Model Development and Hosting: By providing a robust infrastructure for ML model hosting, the platform supports the development and deployment of advanced in-silico models for toxicity prediction. This environment is crucial for the iterative improvement of models and the validation of their predictions.

  3. Gateway for External Access to Project Resources: Beyond the lifespan of the ALTERNATIVE project, the platform aims to serve as a gateway for the broader research community. By exposing data and ML services through a well-defined API, it enables external users to leverage the project's outputs, thereby extending its impact and utility.

Key Objectives of the Cloud Data Platform

  • Facilitation of Data Exchange: A primary goal of the platform is to enable seamless and efficient data exchange between consortium partners throughout the duration of the ALTERNATIVE project. This feature is vital for ensuring that all participants have timely and easy access to the data necessary for the project's success.

  • Hosting of ML In-Silico Models for Toxicity Prediction: The platform serves as a robust environment for hosting advanced Machine Learning (ML) in-silico models. These models are crucial for toxicity prediction, representing a significant stride in the field of predictive analysis and safety assessment. By providing a reliable and scalable hosting solution, the platform ensures that these models are readily accessible and operable for the project's researchers and analysts.

  • Exposing Data and ML Services via API: Post-completion of the ALTERNATIVE project, the platform is designed to extend its utility beyond the consortium. It will expose data and ML services to external users through a well-defined Application Programming Interface (API). This extension will facilitate wider access to the project's valuable resources, thereby enhancing research and development efforts in related fields.

Structure of the Cloud Data Platform

To achieve these ambitious goals, the Cloud Data Platform is constructed as a multi-layered software stack, each layer playing a critical role in the platform's overall functionality and performance:

  • Cloud and Infrastructure Layer: This foundational layer provides the necessary cloud-based infrastructure to support the platform's operations, ensuring scalability, reliability, and security.

  • Administrative Layer: The administrative layer is responsible for overseeing the platform's overall management, including user access control, resource allocation, and monitoring of platform activities.

  • Data Layer: At the core of the platform is the data layer, which handles the storage, processing, and management of data. This layer is optimized for high-performance data operations, essential for the effective functioning of ML models and data exchange processes.

  • APIs Layer: The APIs layer is designed to offer a streamlined and secure interface for accessing the platform's services. It plays a crucial role in integrating the platform with external systems and in making data and ML services available to external users post-project.

  • Microservices Layer: This layer comprises a suite of microservices, each designed to perform specific functions within the platform. The use of microservices architecture enhances the platform's modularity, flexibility, and ease of maintenance.

In summary, the Cloud Data Platform developed for the ALTERNATIVE project represents a sophisticated and multi-faceted solution, tailored to meet the project's unique requirements for data exchange, ML model hosting, and service accessibility. This document will delve into each aspect of the platform, elucidating its design, functionalities, and the value it brings to the ALTERNATIVE project and its stakeholders.