This is a hybrid event. T une in via Webex here.

NLM cell knowledge network: Artificial intelligence approaches for characterizing cell phenotypes from single cell genomics data

If methodological research in bioinformatics can be summarized in a single sentence, it is this: „Our new method outperformed the existing ones“. But is it realistic to expect that every new method outperforms existing ones? Do such claims meaningfully help readers and are they trustworthy? What do bioinformaticians need to identify the most appropriate method for their application? In this talk, I will discuss the importance of neutral method comparison studies as a cornerstone for generating reliable evidence on the performance of methods and providing sound, practical guidance for method selection. Special emphasis will be placed on the design of such studies, the various sources of bias that may affect their results, and the incentive structure in science.Traditionally, a cell’s identity is determined by its morphology and histology characteristics. Recent advancements of single cell genomics technologies have revolutionized our understanding of the diverse cell phenotypes in the human body. Catalyzed by several large international consortia, e.g., the Human Cell Atlas and the BRAIN Initiative, an unprecedented volume of single cell transcriptomic profiling data have been generated, aiming to form a comprehensive landscape of all human cells across organs and biological systems. This presents new opportunities and challenges for the bioscientific research community to incorporate the rapidly growing knowledge about cell phenotypes from the single cell genomics data. At the National Library of Medicine (NLM), we develop the Cell Knowledge Network (CKN) – a public resource that integrates cell type knowledge from single cell experiments with existing knowledge in biomedical ontologies. Cell type knowledge derived from single cell genomics data using validated explainable artificial intelligence algorithms are represented in semantic knowledge graphs using an ontological framework. CKN is interoperable with the Cell Ontology and other reference ontologies, ensuring the provenance and trustworthiness as an AI resource.

About the speaker

Yun (Renee) Zhang is a Principal Investigator in the Division of Intramural Research at the National Library of Medicine, NIH. Her research focuses on developing computational and machine learning methods based on advanced biostatistics to analyze large-scale, multi-omics, single-cell data for identifying disease biomarkers. Her group uses explainable AI approaches and collaborates closely with experimental investigators to characterize cell phenotypes at single-cell resolution. Prior to joining NIH, she was Assistant Professor at the J. Craig Venter Institute. She received her PhD in Statistics from the University of Rochester Medical Center, with additional degrees from the University of Pennsylvania and University of Oxford.