metaphacts, a Digital Science Solution, are delighted to be supporting the 2026 SWAT4HCLS conference as Platinum sponsors.
The conference brings together researchers and industry experts to explore the role of semantic technologies, knowledge graphs and AI in advancing healthcare and the life sciences.
Visit the team at our booth in the exhibit hall to discuss how our solutions help organisations build trusted, explainable and efficient knowledge-driven workflows across the healthcare and pharma landscape and how the Dimensions Knowledge Graph can help achieve knowledge-driven and AI-powered Research and Discovery.
Join us
We also invite you to join our presentation, in collaboration with Systasy:
Semantic Knowledge Graphs for Drug Repurposing and Precision Medicine in Complex Neuropsychiatric Disorders, a collaboration betweeen metaphacts and Systasy
Dr. Sven Wichert
Chief Executive Officer & Co-Founder | Systasy Bioscience GmbH
Samaneh Jozashoori
Senior Technical Consultant | metaphacts
24 March 2026, 13:30 – 14:00
Overview
Drug discovery in neuropsychiatric and other complex diseases is characterized by high failure rates, long development timelines, and limited therapeutic options for many patient groups. A key reason is the pronounced biological and clinical heterogeneity of these disorders, which limits the effectiveness of traditional target-centric drug discovery. Drug repurposing, combined with precision medicine approaches, provides a promising alternative but depends on the systematic integration of heterogeneous biomedical data and experimental validation.
In this presentation, we introduce an end-to-end workflow that combines large-scale biomedical knowledge graphs with omics data and functional assays to support drug repurposing for patient groups with specific genetic predisposition. Scientific publications are integrated into a knowledge graph in which natural-language statements identified in the text are converted into structured, causal relationships between genes, diseases and drugs using large language models. This semantic representation enables transparent querying and reasoning across multiple biological abstraction levels.
Starting from transcriptomic datasets derived from genetically defined disease models, we apply semantic search, filtering, and prioritization to identify repurposable drug candidates that are mechanistically linked to disease-relevant pathways. Candidate compounds are selected based on disease and tissue relevance and published evidence. These candidates are subsequently validated experimentally using multiplexed pathway assays, enabling the assessment of pathway modulation rather than isolated molecular targets.
We demonstrate the approach in a neuropsychiatric disease context with a strong genetic risk background, illustrating how semantic integration supports the identification of compounds that are specifically relevant for distinct molecular subtypes. The results show that knowledge graph–driven workflows can accelerate drug repurposing, reduce experimental search space, and increase the interpretability of discovery decisions.
Overall, this work highlights how Semantic Web technologies can act as a unifying layer between unstructured biomedical knowledge and precision medicine workflows, enabling reproducible, explainable, and patient-centric drug discovery pipelines.
