Sei is a deep-learning-based framework for systematically predicting sequence regulatory activities and applying sequence information to understand human genetics data. The model provides a global map from any sequence to regulatory activities, as represented by 40 sequence classes by integrating predictions for 21,907 chromatin profiles. For more details, find the Sei code repository here or read the manuscript here. See our docs for more details about the web interface.
We've refreshed the visualizations for variant effect predictions (DeepSEA or ExPecto), improving the performance for large sequence submissions (1000+ variants) and visualization of chromosome positions.
ExPecto makes highly accurate cell-type-specific predictions of gene expression solely from DNA sequence (details here). Researchers can now run ExPecto on their variants of interest, along with exploring our pre-computed ExPecto predictions. HumanBase will automatically identify the gene whose TSS is closest to a submitted variant and predict the impact of that variant on the gene's expression in 200+ tissues and cell-types.
By Susan Reslewic Keatley
This powerful, dynamic software system is changing how scientists sift through the recent deluge of human genomic data — and deepening our knowledge of human biology, health and disease. Read more