Broading the Partisipation of ImageAI Scientists in Cryo-ET Analysis

Imaging science is at the forefront of advancements in molecular biology and biomedical technologies. An exciting contribution is electron tomography at cryogenic temperatures (cryo-ET). Cryo-ET offers unparalleled views of life and disease processes at the protein level, opening the door to visual proteomics. Manual analysis of tomograms, although slow and difficult, has revealed insights into biological and pathological processes. But a manual approach is too inefficient to leverage existing (and future) data; image AI is needed. To democratize image AI research and invite fresh perspectives, we aim to broaden the community of scientists and engineers focusing on bacterial structure identification. As an initial step, we are launching a Kaggle competition for predicting the locations of bacterial flagella motors in cryo-ET tomograms. By doing so, we hope the community will develop a deep learning model that can identify structures. We are at an unprecedented time in history where anyone with a computer can contribute to our problem. By encouraging participation from the broader image AI community, we hope to accelerate advancements in this application of imaging science.