Track III Spatiotemporal Omics (Driven by Multi-Tissue Biobanks): Subtrack 5. Plants(J28015)

Foundation Model

Based on BGI’s multi-omics platforms and the Plant Spatiotemporal Omics Consortium (STOC Plant), we construct plant spatiotemporal foundation models. By integrating high-quality multimodal data across the life cycles of Arabidopsis, rice, soybean, and wheat, we aim to achieve digital simulation and functional prediction of organ development and genomic patterns, supporting both smart breeding and fundamental research.

Vertical Applications

• ① Crop Development: Precise simulation of seed and meristem development;
• ② Smart Breeding: Genomic design, parental selection prediction, and environmental perturbation simulation;
• ③ Plant-Environment Interaction: Mechanisms of pathogen infection (e.g., Citrus HLB) and abiotic stress responses;
• ④ Algorithms & Tools: Development of denoising algorithms, small RNA tools, and reads recovery pipelines;
• ⑤ Large-scale Projects: Implementation of the "Wheat Spatiotemporal Atlas Project" and other international initiatives.

Affiliated Branch

Beijing Branch (Beijing)

1. Research Areas

• Plant spatiotemporal omics and integrative multi-omics analysis
• Foundation models for crops and computational biology
• Intelligent breeding and genome design

1. Applicant Background

(Candidates meeting at least one of the following; interdisciplinary backgrounds are strongly encouraged)

• Biological Sciences: Botany, Genetics, Developmental Biology, Agronomy, or related disciplines
• Computational and Quantitative Sciences: Bioinformatics, Computer Science, Artificial Intelligence, Mathematics, Applied Mathematics, or related fields

1. Core Competencies

• Experimental Skills:

Experience with standard molecular biology techniques (e.g., nucleic acid extraction, PCR). Background in tissue sectioning, microscopy, in situ hybridization, or single-cell library preparation is advantageous.

• Bioinformatics and Programming:

Proficiency in Linux and programming in Python and/or R. Experience analyzing high-throughput sequencing data (e.g., scRNA-seq, spatial omics, genomics) is preferred.

• AI and Computational Modeling:

Familiarity with deep learning frameworks (e.g., PyTorch, TensorFlow). Interest or experience in foundation models (e.g., Transformer-based architectures, diffusion models), image analysis, or spatiotemporal data modeling.

• Research Skills:

Strong ability to read scientific literature in English and write academic papers. Capacity for independent thinking and effective collaboration in interdisciplinary teams.

1. Recommended Coursework

• Plant Biology, Plant Physiology, etc.
• Genetics
• Genomics
• Bioinformatics
• Fundamentals of Machine Learning or Deep Learning

Selection Criteria

Preference will be given to candidates who demonstrate strong intellectual curiosity, readiness to engage in frontier interdisciplinary research, and the ability to clearly articulate the rationale, methodology, and contributions of their previous work.