Search for: All records

Graph neural networks (GNNs) are proficient machine learning models in handling irregularly structured data. Nevertheless, their generic formulation falls short when applied to the analysis of brain connectomes in Alzheimer’s Disease (AD), necessitating the incorporation of domain-specific knowledge to achieve optimal model performance. The integration of AD-related expertise into GNNs presents a significant challenge. Current methodologies reliant on manual design often demand substantial expertise from external domain specialists to guide the development of novel models, thereby consuming considerable time and resources. To mitigate the need for manual curation, this paper introduces a novel self-guided knowledge-infused multimodal GNN to autonomously integrate domain knowledge into the model development process. We propose to conceptualize existing domain knowledge as natural language, and devise a specialized multimodal GNN framework tailored to leverage this uncurated knowledge to direct the learning of the GNN submodule, thereby enhancing its efficacy and improving prediction interpretability. To assess the effectiveness of our framework, we compile a comprehensive literature dataset comprising recent peer-reviewed publications on AD. By integrating this literature dataset with several real-world AD datasets, our experimental results illustrate the effectiveness of the proposed method in extracting curated knowledge and offering explanations on graphs for domain-specific applications. Furthermore, our approach successfully utilizes the extracted information to enhance the performance of the GNN. 

Abstract Tribrachidium heraldicumis an Ediacaran body fossil characterized by triradial symmetry. Previous work has suggested that the anatomy ofTribrachidiumwas conducive to passive suspension feeding; however, these analyses used an inaccurate model and a relatively simple set of simulations. Using computational fluid dynamics, we explore the functional morphology ofTribrachidiumin unprecedented detail by gauging how the presence or absence of distinctive anatomical features (e.g., apical pits and arms) affects flow patterns. Additionally, we map particle pathways, quantify deposition rates at proposed feeding sites, and assess gregarious feeding habits to more fully reconstruct the lifestyle of this enigmatic taxon. Our results provide strong support for interpretingTribrachidiumas a macroscopic suspension feeder, with the apical pits representing loci of particle collection (and possibly ingestion) and the triradial arms representing morphological adaptations for interrupting flow and inducing settling. More speculatively, we suggest that the radial grooves may represent ciliated pathways through which food particles accumulating in the wake of the organism were transported toward the apical pits. Finally, our results allow us to generate new functional hypotheses for other Ediacaran taxa with a triradial body plan. This work refines our understanding of the appearance of suspension feeding in shallow-water paleoenvironments, with implications for the radiation of Metazoa across the Ediacaran/Cambrian boundary.