CV
Machine learning researcher and computational biologist. Download the PDF version below.
General Information
| Full Name | Amaya Gallagher-Syed |
| Based in | London, UK |
| a.gallaghersyed@gmail.com | |
| Research | Identifiability and causality in biological systems; training and evaluating single-cell foundation models, with roots in explainable computational pathology. |
| Languages | English, French, Spanish (all native) |
Research Experience
-
2026 (incoming) Eric & Wendy Schmidt AI for Science Fellow
Imperial College London (I-X) - Independent research programme on identifiable representation learning for biology foundation models, with applications in dynamical systems and cross-modal causal discovery.
-
2025 - present Postdoctoral Research Fellow
Queen Mary University of London · Imperial College London - Mechanistic probing of single-cell foundation models via differentiable gene regulatory network recovery from CRISPR Perturb-seq data.
- Explainable deep learning for cancer survival prediction.
- Supported by a Wellcome Trust Early Career Transition award.
-
2021 - 2025 Wellcome Trust Doctoral Researcher
Queen Mary University of London - Developed explainable, multimodal deep learning for computational pathology in autoimmune disease and cancer, spanning graph neural networks, attention-based multiple-instance learning, and image–omics fusion.
-
2021 - 2023 Teaching Assistant
Queen Mary University of London - Statistics for AI and Data Science · Post-Genomics Bioinformatics.
-
2018 - 2019 Teaching Assistant
Universidad de Buenos Aires - Botany & Plant Physiology.
Education
-
2020 - 2025 PhD in Science (Wellcome Trust)
Queen Mary University of London - Thesis: Explainable Deep Learning for Multi-Stain and Multimodal Computational Pathology in Autoimmune Diseases.
- Supervisors: M. R. Barnes · M. J. Lewis · G. Slabaugh.
-
2019 - 2020 MSc in Mathematical Sciences
Queen Mary University of London - Distinction.
- Thesis: Machine learning for gene expression data of early rheumatoid arthritis patients to distinguish treatment responders from non-responders.
-
2012 - 2019 Licentiate in Computational Biology
Universidad de Buenos Aires - 8.3/10 (Honours). Seven-year full-time degree spanning biology, chemistry, physics, mathematics, statistics, and programming.
Funding & Awards
-
2026 - Eric & Wendy Schmidt AI for Science Postdoctoral Fellowship (≈ £120k)
- Outstanding Reviewer, CVPR 2026
-
2025 - Wellcome Trust Early Career Transition Fund (≈ £40k)
-
2024 - Top Reviewer (top 4%), NeurIPS 2024
-
2020 - Wellcome Trust PhD Studentship in Science (≈ £200k)
Invited Talks
-
2025 Keynote, LatinX in AI Workshop @ ICML
- From Buenos Aires to London: how broad scientific foundations shaped my AI research.
-
2025 AI × Bio Conference, Wellcome Connecting Science
- BioX-CPath: Biologically-driven Explainable Diagnostics for Multistain Computational Pathology.
Technical Skills
-
Machine learning
- Representation learning, foundation-model probing & benchmarking
- Transformers, graph/hypergraph neural networks, multimodal fusion, computer vision
-
Modeling
- Differentiable programming (implicit differentiation), neural ODEs & dynamical systems
- Causal inference from interventional data, gene regulatory network inference
-
Programming & infrastructure
- Python (PyTorch, PyTorch Geometric/DGL), Scanpy/AnnData, Git
- Multi-GPU / distributed training (SLURM, HPC)
-
Biology
- Single-cell & spatial transcriptomics, CRISPR Perturb-seq
- Immunology, systems biology, computational pathology
Academic Service
- Reviewer: NeurIPS 2026, ECCV 2026, CVPR 2026, ICLR 2026, TMLR, CVPR 2025, ICLR 2025, NeurIPS 2024, MICCAI 2023.
- Open source: contributor to the Graph Structure Learning Benchmark (NeurIPS 2023).
- Organiser: DERI Lunch & Learn Seminar. Co-organiser: Responsible AI Workshop.