Dr Eli Niktab
Content navigation
About
Dr Eli Niktab is Principal AI Engineer at the National Centre for Indigenous Genomics (NCIG), John Curtin School of Medical Research, where he leads the architecture and development of AI systems for genomic analysis. His work focuses on developing machine learning and graph based methods for representing, assembling and interpreting human genomes with applications in disease genetics, pharmacogenomics and precision medicine.
Before joining ANU in 2024, he spent a decade bridging applied research and production engineering to deliver scalable machine learning systems across industry, government, and academia. At Callaghan Innovation, New Zealand's former national innovation agency, he served as a core engineer and technical lead, driving the Precision Medicine initiative in collaboration with the country's top scientists. His portfolio spanned proprietary petabyte-scale GPU-accelerated genomic search systems, highly optimised inference for edge devices, graph-based machine learning, and computer vision methods. Previously, at the New Zealand Ministry of Education (2019–2021), he led the development of a first-of-its-kind natural language processing suite for te reo Māori.
He received his PhD from Victoria University of Wellington in 2023. His doctoral work developed graph neural networks, representation-learning methods and sequence models for analysing whole-genome data from more than 70,000 individuals using distributed high-performance computing. The work investigated genomic modifiers of disease onset and severity, including Huntington's disease and Niemann-Pick disease type C, and combined large-scale genomic computation with graph representations of biological systems.
His earlier training in the Munkacsi and Atkinson laboratories centered on high-throughput experiments and lab-in-the-loop methodologies, integrating automated synthetic genetic array (SGA) screening, robotic live cell imaging, and custom image processing algorithms for yeast colonies. Through this pipeline, the team identified genetic interactions modifying disease severity in a yeast model of Niemann-Pick disease type C and mapped the statin genetic interaction network.
He welcomes enquiries from Honours, Masters, and PhD students interested in computational genomics and machine learning.
Affiliations
Research interests
- Genomic AI and foundation models: DNA language models, generative models and deep learning for learning directly from genomic sequence
- Graph and manifold learning: representation of genomes, regulatory networks and biological systems as structured graphs and manifolds
- Genome functional annotation: prediction and interpretation of the functional consequences of genomic variation
- Pangenomes and genome graphs: computational methods for representing genetic diversity at scale
- High-performance genomic computing: GPU-accelerated algorithms and distributed computing for large-scale genomic analysis
- Governance-aware AI: reproducible machine learning and MLOps for controlled genomic data
Teaching information
Student supervision
Students from computer science, statistics, mathematics, bioinformatics or biology with strong quantitative and programming interests are encouraged to get in touch. Python is expected; experience with C++/CUDA, PyTorch, JAX, TensorFlow or distributed/HPC environments is advantageous.
To enquire about supervision, email eli.niktab@anu.edu.au with a short description of your research interests, relevant technical or biological experience, a list of publications and the type of research problem you would like to explore.