Chapter 1 Deploying, Training, and Interpreting Deep Learning Models for regulatory genomics in AnVIL

Led by: Anshul Kundaje & Vivekanandan Ramalingam, Stanford University

AnVIL Outreach coordinator: Kate Isaac

1.1 About This Track

The “Deploying, Training, and Interpreting Deep Learning Models for Regulatory Genomics in AnVIL” CoFests! track at the AnVIL Community Conference offers hands-on training for users and developers interested in applying deep learning to regulatory genomics. This track aims to demonstrate how deep learning models can be utilized for functional genomics datasets, such as ChIP-seq and ATAC-seq, at scale via AnVIL. Participants will develop a comprehensive understanding of the steps involved in deploying, training, and interpreting these models, including the available input options and how to leverage the resulting outputs to address various biological questions.

1.2 Workspaces

The workspaces that will be used for this CoFest! are used to train and analyze base pair resolution neural network models on Transcription factor ChIP-seq datasets (BPNet) and to process train and analyze ChromBPNet style models on ATAC and DNase datasets.

1.3 Report Out

At the conclusion of the CoFest!, we will spend some time to collaboratively develop a user guide for utilizing these deep learning workspaces. This user guide will be published on GitHub.