Release Notes

This page documents the changes introduced in each version of the UniVerse Platform.

Version 1.7.0 (2025-08-29)

Notes MVP Implementation

We're excited to introduce the Notes MVP feature, a comprehensive note-taking system integrated throughout the platform. This release includes both frontend and backend implementation, allowing users to create, manage, and organise notes seamlessly within their workflows.

ChatMDI & Notes Integration

Enhanced the ChatMDI interface with full Notes integration, enabling users to save conversation insights and reference materials directly from chat interactions. This streamlined workflow improves productivity by connecting conversational AI with persistent note storage.

LLM-Powered Notes Enhancement

Integrated a lightweight LLM to provide intelligent assistance within the Notes feature, including content suggestions and automated categorisation. This AI-powered functionality helps users organise and enhance their note-taking experience.

Database Migration: Neo4j to PostgreSQL

Completed migration of MDI components from Neo4j to PostgreSQL for improved performance and maintenance efficiency. This 12-hour maintenance window transition provides better scalability and reduces infrastructure complexity.

Enhanced Monitoring Stack

Deployed comprehensive monitoring solutions covering both infrastructure and application layers. These monitoring improvements provide better visibility into system performance, enabling proactive issue detection and resolution.

MongoDB Data Consolidation

Please note that during the MongoDB consolidation process affecting workflow, runs, and datasets, there is a risk of data loss. Users are advised to backup critical data before the maintenance window. Our team will work to minimise any impact, but please plan accordingly.

Version 1.6.0 (2025-08-07)

Enable a Run to be created and execute a workflow

Users can now create and launch workflow runs directly through the platform interface. This feature streamlines the process of executing automated workflows.
This release represents the first version of the feature and is being made available in an early-stage, experimental capacity. The goal of this initial implementation is to verify that the key components of the workflow system are successfully integrated and able to communicate — including the user interface, orchestration layer, and execution backend.
While the core functionality is in place, this version is best viewed as a technical foundation, and we expect to expand, refine, and harden the feature in future iterations

Visualise Runs Progress and Basic Logging

The system now provides real-time visual tracking of workflow run progress with integrated logging capabilities. Users can monitor execution status and view basic log outputs to troubleshoot and track workflow performance.

Limit Cached Workflows to Supported Languages

Workflow caching has been restricted to only languages that are officially supported by the platform. This change ensures better reliability and prevents caching issues with unsupported language environments.

Fix volumes issue after the new CSI driver in TES

Volume mounting problems that occurred following the TES CSI driver update have been resolved. This fix restores proper data persistence and storage functionality for containerised workloads.

Transfer pacificanalytics.com to Cloudflare

The pacificanalytics.com domain has been migrated to Cloudflare's infrastructure for improved performance and security. This transition provides enhanced CDN capabilities for the resources hosted under the domain, including the UniVerse platform.

Openstack Monitoring enhancement

Monitoring capabilities for the OpenStack infrastructure have been upgraded with additional metrics and alerting features. These improvements provide better visibility into system health and performance across the cloud environment.

Clean up the API Gateway to support the new architecture

The API Gateway has been refactored and optimised to align with the updated system architecture. Legacy components have been removed to improve performance and maintainability of the gateway services.

Save and Export Cohorts

Users can now save cohort definitions for reuse and export cohort data for external analysis. This functionality enables better data management and facilitates sharing of research datasets across different platforms.

Version 1.5.0 (2025-07-08)

Cohort Builder Front-End for create and visualise a Cohort

The PA Cohort Builder now features a fully functional front-end, empowering users to intuitively create and visualise biological sample cohorts. This user-friendly interface simplifies the complex process of leveraging the underlying knowledge graph for cohort definition and analysis. A comprehensive user guide has also been developed to ensure a smooth and effective user experience.

Backend API(s) For Cohort Builder

The foundational backend APIs for the PA Cohort Builder are now complete. These APIs are crucial for programmatically interacting with the system's underlying Neo4j knowledge graph, enabling robust construction and querying of biological sample cohorts. This achievement lays the groundwork for the platform's advanced functionalities.

Streamline Data ETL To Cohort Builder

Our data ETL (Extract, Transform, Load) processes for the PA Cohort Builder have been significantly streamlined. This optimisation ensures more efficient and reliable ingestion of heterogeneous biomedical data into the knowledge graph. Users will experience improved performance and data availability for building and managing cohorts.

Enhance code quality for MDI ingestion and backend endpoints

Comprehensive unit tests have been implemented for both MDI (Medical Device Information) data ingestion and all backend endpoints of the Cohort Builder. This robust testing regime ensures the integrity and accuracy of data flowing into the Neo4j knowledge graph. It significantly enhances the overall stability and reliability of the platform for biological sample cohort management.

Sample Sheet for nf-core RNAseq workflow (nf-core/rnaseq)

A standardised sample sheet has been created specifically for the nf-core RNAseq workflow. This addition will streamline the preparation of biological sample data for RNA sequencing analysis, ensuring compatibility and ease of use with the nf-core/rnaseq pipeline. It aims to enhance efficiency and reduce common errors in data processing for researchers.

Version 1.4.0 (2025-06-10)

Advanced Workflow Management

The platform now features comprehensive GA4GHTool Registry Service (TRS) API support with a unified workflow browsing interface and dynamic filtering capabilities. With these, users can conveniently discover and access public workflows in a standards-compliant manner. Uploading user workflows now offers various annotation, content ingestion and editing options that ensure seamless integration with the new workflow discovery and use functionalities.

Public Repository Integration

With the integration of WorkflowHub as an additional workflow source next to Dockstore (see previous release), the platform now supports two of the most comprehensive workflow registries in the field. Users can now browse, search, and import publicly available workflows directly from these synchronised repositories, significantly expanding the available toolkit for research and analysis.

Enhanced AI and Multi-Agent Systems

The platform's AI capabilities have been significantly upgraded with updated agent architecture and multi-agent system integration. The user experience has also been enhanced with the adoption of streamed responses, making messages appear in real time as they are generated.

Version 1.3.0 (2025-05-13)

Bioinformatics Tools Integration

The platform now supports integration with external workflow registries, allowing users to browse and import bioinformatics workflows straight into the platform. Users can search and filter tools by class and language, view comprehensive tool details, and import tools with a single click. Currently integrated registries include Dockstore, with more registries planned for future releases.

Support for deploying imported workflows is also planned for a future release.

Version 1.2.0 (2025-04-11)

Hypotheses Integration

An alpha version of a chatbot for generating scientific hypotheses is now integrated into the platform. It provides AI assistance to researchers for establishing new research directions based on supporting literature evidence. The current version only considers a limited number of abstracts of scientific publications and a simple agentic workflow for hypothesis generation.

Minor UI Updates

Pipelines has been renamed to "Runs" to be more consistent with industry terminology. The functionality of the module has not changed. Other labels have been updated to keep tense consistent throughout the platform.

Live Online Documentation

The online user documentation has been updated to follow UniVerse's branding in terms of wording, logo use and colour theming.

Version 1.1.0 (2025-04-01)

JupyterHub Integration

JupyterHub enables multiple users to run, manage, and share Jupyter Notebooks on a centralised server, providing a collaborative, scalable, and customisable environment for data science, education, and research.

Sign In via Keycloak

Signing In to the platform is now done via Keycloak. Keycloak is an open-source security software tool that provides secure and centralised authentication, and manages identities and access control for applications and services.

Live Online Documentation

Online user documentation is accessible from the navigation sidebar in the platform, with real-time updates, improved accessibility, and search functionality. Stay informed with the latest changes as they are made available.

Integration of New Backend

The platform now includes an enhanced backend integration with an API Gateway, improving stability, scalability, and facilitating seamless interaction with other services.

UniVerse Branding

The platform has been updated to include UniVerse logos and colour theming.

Terms of Service

The login page now includes a link to Terms of Service.

Version 1.0.0 (2025-03-12)

Prototype Version

This version of the platform shows the conceptual functionality of the Datasets, Workflows and Pipelines modules. While each module is functional and interactive, we recommend limiting the sise of files uploaded to less than 4GB. Additionally, the current compute power of these modules is suitable for proof of concept use cases only.

Datasets

Use the Datasets functionality to securely store your data in the Platform so that you can then run a Workflow on it.

Workflows

Workflows let you analyse your data by following a predefined set of steps. A Workflow is executed on a dataset in a Pipeline. The Platform currently supports workflows written in the Nextflow "workflow definition language".

Pipelines

Pipelines let you configure and run individual Workflows on Datasets.