Machine learning engineer and data scientist with experience across healthcare, reinsurance, and construction, specializing in computational biology and computer vision. Develops methods for spatial omics data in cancer research and builds end-to-end ML pipelines and open-source tools, translating research into real-world applications. Familiar with drug discovery workflows and pharmaceutical R&D, with a focus on connecting machine learning to biomedical discovery.

• Machine Learning & Data Science: Computer Vision · Self-Supervised Learning · Representation Learning · Multi-Modal Modeling · Statistical Analysis
• Tooling & Engineering: PyTorch · Lightning · scikit-learn · TensorFlow · Git · Snakemake · Ray · Docker · FastAPI · Azure ML Studio · Azure DevOps · W&B
• Domain Expertise: Spatial Omics · Single-Cell Analysis · Digital Pathology · Drug Discovery Workflows
• Languages: German (native) · English (fluent) · French (beginner) · Spanish (beginner)

• Multi-Modal Representation Learning for HnE and Xenium - A novel early-fusion model for multi modal representation of Xenium and HnE modalities.
• OmicsEmbed: A standardized data processing and self-supervised learning framework for highly multiplexed tissue images, with a focus on imaging mass cytometry.
• ATHENA — Python toolkit for representation learning and statistical analysis of spatial single-cell data.
• SpatialProteomicsNet — Unified data access layer for spatial omics datasets.
• DEL-Hit — Framework for analyzing DNA-encoded chemical libraries with high-throughput performance.

➡️ Explore the full list on GitHub.

• A. Martinelli, B. Illing, I. Katircioglu, et al. Learning Joint Morpho-Molecular Tissue Representations with a Multimodal Transformer. ICLR, 2026.

• A. Martinelli, M. Rapsomaniki. SpatialProteomicsNet: unified access to spatial proteomics datasets. Journal of Open Source Software, 2026.

• A. Martinelli, A. Lessing, G. Hoppeler, et al. DEL-Hit: a computational framework for DNA-encoded libraries. Under revision at Nature Protocols, 2026.

• A. Martinelli, F. Bonollo, S. Karkampouna, et al. Cellular and molecular profiling of the prostate cancer microenvironment. Under review at Nature Communications, 2026.

• P. Pati, S. Karkampouna, F. Bonollo, ..., A. Martinelli, et al.
  Accelerating histopathology workflows with generative AI-based virtually multiplexed tumour profiling. Nature Machine Intelligence, 2024.

• M. Keller, D. Petrov, A. Glogger, ..., A. Martinelli, et al.
  Highly pure DNA-encoded chemical libraries by dual-linker solid-phase synthesis. Science, 2023.

• A. Martinelli, M. A. Rapsomaniki.
  ATHENA: analysis of tumor heterogeneity from spatial omics measurements. Bioinformatics, 2022.

• A. Martinelli, J. Wagner, B. Bodenmiller, et al.
  scQUEST: scQUEST: Quantifying tumor ecosystem heterogeneity from mass or flow cytometry data. STAR Protocols, 2022.

🔍 See more on Google Scholar.