Harmonizome 3. 0: A Giant Leap in Understanding Genes and Proteins

You might be wondering what makes the Harmonizome 3. 0 so special. Well, let’s dive into it! This new version of the Harmonizome database has been upgraded to include a whopping 26 new datasets. What does that mean? It means that it now contains nearly 12 million associations between genes and various attributes like types of cells and tissues, diseases, and pathways. That’s a lot of data to help scientists understand our genes better! But wait, there’s more! This upgrade comes with a nifty feature called dataset crossing. What’s that, you ask? It’s a way to find gene modules that are shared across these datasets. Pretty cool, huh? And if you’re wondering why some datasets have a high gene set overlap, don’t worry! A large language model is here to explain it by composing a paragraph that speculates about the reasons.

Not just that, but Harmonizome 3. 0 also offers more data formats and visualization options. You can download the datasets as knowledge graph (KG) assertions and even visualize them using Uniform Manifold Approximation and Projection (UMAP) plots. Plus, you can explore these KG assertions through a user interface that shows gene-attribute associations as ball-and-stick diagrams. Overall, Harmonizome 3. 0 is like a treasure trove of processed omics datasets, ready to be used with AI. It’s available online at https://maayanlab. cloud/Harmonizome/. So, if you’re into genes and proteins, this is a resource you don’t want to miss!

Actions