Decoding gene regulatory networks in space

• Applicant: Professor Dr. Ivan Gesteira Costa Filho
• Subject Area: Bioinformatics and Theoretical Biology
• Term: since 2017
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 388802535

Spatial multimodal genomics provides a unique technology to understand chromatin and transcriptional changes governing spatially constrained cell differentiation processes, including organogenesis and tissue regeneration. However, integrating multimodal single data is challenging due to modality-specific properties. Existing computational approaches overlook spatial dimensions, and the inference of gene regulatory networks linked to spatially distributed expression gradients remains unexplored. This proposal aims to address these gaps through two main aims. First, we will propose a spatially aware canonical correlation analysis (CCA) to integrate RNA, open chromatin, and histology data. For this, we will investigate the use of linear and non-linear CCA approaches to capture complex cell differentiation patterns. These models will provide latent space representations, which can be used for segmentation and detection of spatial domains. Moreover, we will explore the interpretability of canonical components, i.e., associating them with molecular features related to the detected spatial domains. Next, we will propose computational methods for the inference of spatially driven differentiation gene regulatory networks. This includes the delineation of regulatory features, such as chromatin accessibility, transcription factor activity, enhancer-to-gene links, associated with development gradients by leveraging denoising properties of CCA methods. This will allow the inference of spatially resolved gene regulator networks to find TFs, regulatory regions, and expression programs associated with cell differentiation or disease-related cellular transformations. These methods will be used to characterize regulatory mechanisms associated with tissue remodeling in kidney diseases.

• DFG Programme: Research Grants