Understanding the genetic factors that underlie the normal variation in cardiac anatomy is of great interest. In this study, Rodrigo Bonazzola et al. applied unsupervised geometric deep learning to phenotype the left ventricle using an MRI-derived three-dimensional mesh representation (as depicted on the cover). We show that this approach boosts genetic discovery and provides deeper insights into the genetic underpinnings of cardiac morphology.
Check out https://lnkd.in/edrvTg2W
The ClinGen Pharmacogenomics Working Group: Developing frameworks for evaluating pharmacogenomic gene validity and actionability
Read it here: https://www.sciencedirect.com/science/article/pii/S294977442400222X
Generating a framework for curating mechanism of disease in monogenic conditions: A consensus effort of the Gene Curation Coalition*
Read it here: https://www.gimopen.org/article/S2949-7744(24)00622-8/fulltext
DBDS PhD student Yusuf Roohani’s work on universal cell embeddings was featured in a New York Times article: “A.I. is learning what it means to be alive.” This foundation model for cell biology is able to map any cell from any species or tissue into a unified latent space, eliminating the need for model tuning or label information. The article describes how such efforts are bringing scientists closer to the vision of a virtual cell.
Congratulations, Yusuf!