Presented By: DCMB Seminar Series
DCMB / CCMB Weekly Seminar featuring Christina Leslie, PhD (of Memorial Sloan Kettering Cancer Center)
Advances in predictive models for single-cell and regulatory genomics
Abstract:
The last several years have brought notable successes in the application of machine learning approaches, and especially deep learning models, to problems in single-cell and regulatory genomics. The advent of single-cell chromatin accessibility (scATAC-seq) and multiome (scRNA+ATAC-seq) brings new machine learning challenges and opportunities to link chromatin state to developmental trajectories, gene regulation, and even higher order chromatin organization. We will present recent models from our group to exploit these new single-cell data modalities: CellSpace, a sequence-informed embedding algorithm for scATAC-seq that learns biologically meaningful latent structure while mitigating batch effects; SCARlink, a gene-level regression model for multiome data that identifies cell-type-specific enhancers and enables interpretation of disease-associated genetic variants; and ChromaFold, a deep learning model that predicts the 3D contact map from scATAC-seq alone.
https://umich-health.zoom.us/j/93929606089?pwd=SHh6R1FOQm8xMThRemdxTEFMWWpVdz09
The last several years have brought notable successes in the application of machine learning approaches, and especially deep learning models, to problems in single-cell and regulatory genomics. The advent of single-cell chromatin accessibility (scATAC-seq) and multiome (scRNA+ATAC-seq) brings new machine learning challenges and opportunities to link chromatin state to developmental trajectories, gene regulation, and even higher order chromatin organization. We will present recent models from our group to exploit these new single-cell data modalities: CellSpace, a sequence-informed embedding algorithm for scATAC-seq that learns biologically meaningful latent structure while mitigating batch effects; SCARlink, a gene-level regression model for multiome data that identifies cell-type-specific enhancers and enables interpretation of disease-associated genetic variants; and ChromaFold, a deep learning model that predicts the 3D contact map from scATAC-seq alone.
https://umich-health.zoom.us/j/93929606089?pwd=SHh6R1FOQm8xMThRemdxTEFMWWpVdz09
Related Links
Explore Similar Events
-
Loading Similar Events...