The Society of Biomolecular Imaging and Informatics Annual Conference
Boston, MA, USA
October 27–29, 2025
Meet the Genedata experts at the Biomolecular Imaging and Informatics 2025 in Boston, MA, USA at booth #23.
Genedata Screener® automates assay analysis, validates raw data and assay result quality, and consolidates assay information across the enterprise and external collaborators. By integrating and condensing complex experimental data, it powers project decisions and lays the foundation for more predictive, AI-driven, drug discovery and development.
Don't miss the opportunity to see how Genedata Screener® for HCS. The world’s top pharma and contract research organizations rely on Screener for HCS for a streamlined high content analysis. Screener manages massive, multi-featured HCS data, and uncovers relevant features with powerful analysis methods.
If you would like to schedule a meeting in advance or receive more information on Genedata Screener, please contact screener(at)genedata.com.
Recommended Poster Presentation
Unsupervised Phenotype Discovery in High-Content Imaging via Archetypes and Self-Supervised Learning (#2)
Mario Wieser, Daniel Siegismund, Stephan Steigele | Genedata AG, Basel, Switzerland
Identifying novel therapeutic candidates for complex diseases remains a majorchallenge in modern drug discovery. To address this, biopharmaceutical researchincreasingly relies on automated, high-throughput screening assays using cell culturemodels to evaluate thousands of compounds in parallel. However, the resulting large-scale imaging data complicates systematic expert review, making phenotype discoveryand classification dependent on extensive—and often biased—manual curation.
A common strategy to mitigate this issue is archetypal analysis, which identifiesphenotypes within a dataset. In this work, we introduce an end-to-end deep learningframework that simultaneously learns embeddings from high-content images anduncovers phenotypic structures without supervision [1]. Building on theserepresentations, we apply self-supervised learning to construct a phenotypic embeddingspace, enabling intuitive visual exploration and downstream assay analysis.
Comprehensive experiments on industry-relevant assays demonstrate that ourapproach outperforms existing unsupervised and supervised methods, providing ascalable and unbiased pipeline for drug screening and functional genomics [2].
[1] Wieser, M et al. "Revisiting Deep Archetypal Analysis for Phenotype Discovery in High Content Imaging."2025 IEEE/CVF WinterConference on Applications of Computer Vision (WACV). IEEE Computer Society, 2025.
[2] Siegismund, D et al. "Self-supervised representation learning for high-content screening."International Conference on MedicalImaging with Deep Learning. PMLR, 2022
Agentic AI Enables Automated CellProfiler Pipeline Design via Multimodal In-Context Learning (#3)
Daniel Siegismund, Mario Wieser, Cameron Scott, Stephan Steigele | Genedata AG, Basel, Switzerland
Designing image analysis pipelines in CellProfiler typically demands expert knowledgeand extensive manual tuning, creating barriers to scalability and accessibility in high-throughput biological research. In this proof-of-concept study, we introduce an agenticworkflow that harnesses in-context learning and multimodal AI models to automaticallygenerate CellProfiler pipelines from minimal user input.Our approach enables a vision-language model to function as an autonomous agent:interpreting example images and textual prompts, reasoning over visual features, andproducing complete pipeline configurations tailored to new datasets. Leveraging a few-shot learning paradigm, the system generalizes from a single annotated example—comprising an image, text, and documentation context—to generate parameterizedpipelines for tasks such as segmentation and object detection, adapted to the biologicalcontext presented.Quantitative evaluations demonstrate that the AI-generated pipelines achieveperformance comparable to expert-designed configurations while reducing complexity.These findings highlight the potential of agentic AI to streamline and democratizebiomedical image analysis by minimizing domain-specific expertise requirements andsimplifying pipeline design.
