*At the end of this competition, we will write a scientific article describing this event. If you would like to be considered as a co-author, please opt-in in [this form](https://forms.gle/LX5tDUqqGuN7KL9D8). The organizers reserve the right to select co-authors based on contribution merit.*

With every breath, meal, and workout, your body sends signals using complex protein interactions. Nearly every biological process, like transporting oxygen and building muscle mass, relies on proteins. However, we still don’t fully understand what many individual proteins do. Learning how proteins function is key to understanding how our cells work and creating new disease treatments.

Current approaches for understanding a protein’s function are to compare it with proteins whose functions are known, search scientific papers for clues, or use machine learning to analyze data from multiple sources. While these methods are helpful, they still struggle with complex biology. Proteins often have multiple roles, and their functions can change depending on the situation.

In this competition, you’ll train a model to predict Gene Ontology (GO) terms for a set of proteins based on their amino acid sequences. These GO terms describe what the protein does, which biological processes it’s involved in, and where in the cell it operates. You’ll need to predict multiple labels at once while sorting through messy biological data.

If your model performs well, it could help scientists narrow down protein roles faster, prioritize lab experiments, and spot connections that might otherwise go unnoticed.  Knowing what a protein does is a small step for a model, but a giant leap for medicine.