YoVDO

Predicting Single-Cell Perturbation Responses for Unseen Drugs

Offered By: Valence Labs via YouTube

Tags

Drug Discovery Courses Machine Learning Courses Genomics Courses Transfer Learning Courses RNA-Seq Courses Encoder-Decoder Architecture Courses

Course Description

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a comprehensive conference talk on predicting single-cell perturbation responses for unseen drugs. Delve into the world of single-cell transcriptomics and its role in studying cellular heterogeneity. Learn about the challenges of scaling high-throughput screens and the importance of transferring information from bulk RNA-seq data. Discover a new encoder-decoder architecture and transfer learning scheme designed to improve generalization performance and reduce the need for extensive single-cell screens. Gain insights into the potential of this method for generating in-silico hypotheses and accelerating targeted drug discovery. Follow along as the speakers discuss cellular underpinnings of health and disease, single-cell genomics, perturbation modeling objectives, and the introduction of chemCPA. Explore datasets, transfer learning strategies, and experiments benchmarking molecule encoders. Conclude with an examination of chemCPA for unseen drugs and participate in a Q&A session.

Syllabus

- Intro
- Understand the Cellular Underpinning of Health & Disease
- Single-Cell Genomics
- Objectives for Perturbation Modelling
- Introducing chemCPA
- Datasets
- Transfer-Learning for chemCPA and Evaluation Strategy
- Experiments: Benchmarking Molecule Encoders
- chemCPA for Unseen Drugs
- Conclusion
- Q+A


Taught by

Valence Labs

Related Courses

Bioinformatic Methods I
University of Toronto via Coursera
Bioinformatic Methods II
University of Toronto via Coursera
Statistics for Genomic Data Science
Johns Hopkins University via Coursera
Genomic Technologies in Clinical Diagnostics: Next Generation Sequencing
St George's, University of London via FutureLearn
Data Analysis for Genomics
Harvard University via edX