Innovative AI-empowered organoid platform for illuminating early neural tube development and related neural tube defects

• Grant number: 24/10218-8
• Support Opportunities: Regular Research Grants
• Start date: February 01, 2025
• End date: January 31, 2028
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Oswaldo Keith Okamoto
• Grantee: Oswaldo Keith Okamoto
• Principal researcher abroad: Guang Yang
• Institution abroad: Imperial College London, England
• Host Institution: Instituto de Biociências (IB). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Chunling Tang ; Ma Hui Ling ; Mayana Zatz

Abstract

Neural tube defects (NTDs) are severe congenital anomalies that impact the central nervous system (CNS), posing significant public health challenges, especially in Brazil. These defects originate from malformations during the early stages of neural tube (NT) development. Despite advancements in understanding neural stem cell behavior and neurogenesis, substantial knowledge gaps persist due to the inaccessibility of human tissue and ethical considerations associated with studying human embryonic development. Traditional research methods, such as cell lines and animal models, often fail to replicate the complex 3D architecture and intricate developmental mechanisms of the human CNS. In recent years, human organoids, derived from induced pluripotent stem cells (iPSCs), have emerged as a transformative tool in CNS research. These organoids closely mimic human-specific conditions and the complex 3D structure of the CNS, and can prove valuable insights into neural differentiation and NTDs. Early neural tube organoid models replicate the initial stages of NT formation, offering a unique opportunity to study neural progenitor behavior and cellular environments during critical development stages. These models hold significant potential for identifying therapeutic targets and elucidating the molecular mechanisms underlying NTDs. Fluorescence microscopy assays are indispensable for investigating cellular responses in CNS-related organoid models, offering detailed visualization of cellular structures and interactions. However, these techniques are often labor-intensive and time-consuming. The integration of artificial intelligence (AI) into biomedical research has revolutionized image analysis, enhancing accuracy and interpretation. Advanced AI techniques, such as convolutional neural networks (CNNs) and generative AI models, including variational autoencoders (VAEs) and generative adversarial networks (GANs), have significantly advanced the field of microscopy-based imaging analysis in organoid research.This project aims to leverage organoid technology, high-resolution imaging, and AI to comprehensively explore early neural tube development and related defects. The primary objectives of the project include developing a novel organoid culture platform, implementing a foundation model-assisted approach for extracting morphological features from microscopy images, and developing a generative AI-based approach for modeling early NT development and the pathological progression induced by NTD-related mutations.To achieve these objectives, we will develop a novel organoid platform by creating a high-throughput, low-variability culture system with easy imaging accessibility. By introducing NTD-related mutations in hiPSCs via CRISPR-Cas9, we will study the impact on NT organoid morphogenesis using RNA sequencing, qPCR, and immunocytochemistry. Foundation models will be used to automate the detection and segmentation of organoids in microscopy images. This involves collecting and preprocessing image data, comparing segmentation models, and extracting morphological features. Advanced feature decomposition techniques will be used to analyze differences between NT organoids from isogenic controls and those with NTD mutations. Furthermore, a conditional latent diffusion model will be developed to generate synthetic immunofluorescence (IF) images from brightfield images and RNA sequencing data. This AI model will simulate the dynamics of different cell subsets during NT organoid development, enabling the study of how NTD-related mutations affect early differentiation processes. The integration of organoid technology, advanced imaging, and AI in this project is expected to provide a comprehensive model for observing and analyzing early NT development and related defects. Ultimately, this project aims to improve prevention strategies and therapeutic interventions for NTDs, contributing to better healthcare outcomes in Brazil. (AU)