Post Doctoral Scholar - Michelmore Lab

Postdoctoral Scholar in AI and Foundation Models for Plant Genomics

The UC Davis Genome Center is recruiting a Postdoctoral Scholar to work on a collaboration between the laboratories of Richard Michel more (Genome Center), Xin Liu (Computer Science), and Christine Diepenbrock (Plant Sciences). This project will develop one of the first crop-specific multimodal foundation models integrating more than 100 telomere-to-telomere lettuce genomes, population-scale genomic variation, transcriptomics, and extensive phenotypic datasets to predict the consequences of allelic variation, genome editing, and genotype-by-environment interactions.

The focus of this work will be on the fine-tuning, evaluation, and multi-faceted deployment of a foundation model for lettuce. An existing DNA foundation model architecture will be leveraged while also incorporating advances due to the rapid evolution of the DNA and other
-omic foundation model space. The extensive existing data sets will be leveraged for training, evaluation, validation, and use cases. The project emphasizes reproducible research and open-source software development. The successful candidate will have opportunities to publish both methodological advances in AI and biological discoveries enabled by the models.

• Fine-tune a pretrained foundation model using lettuce genome data with applications in crop improvement.
• Implement appropriate strategies to optimize model performance and benchmark and compare models.
• Collaborate closely with other project team members who have expertise in lettuce genomic resources, remote sensing, plant physiology, and development to
  1) curate training data, including multi-omic and phenotypic data; and
  2) generate hypotheses for model training.
• Supervise undergraduate researchers with training in machine learning.
• Publish findings in machine learning and computational biology journals.

• Ph.D. in Computer Science, Computational Biology, or a related field
• Experience programming in Python
• Demonstrated experience developing, training, or adapting deep learning models
• An interest and willingness to learn about genome biology, gene function, and regulatory circuits
• Willingness to collaborate in multi-disciplinary teams
• Evidence of research productivity through publications in machine learning, computational biology, genomics, or related areas