Hello AIRG,
Next week I will discuss an application of modern machine learning methods in an exciting and emerging area of computational biology: single-cell RNA sequencing analysis.
Title: Bayesian Inference for a Generative Model of Transcriptome Profiles from Single-cell RNA Sequencing â
Authors: Romain Lopez and others â
Paper:
https://doi.org/10.1101/292037â
Presenter: Matt Bernstein
âTime: 4pm - Oct 3, 2018â
Place: CS 3310
Summary:â
Single-cell RNA-sequencing is an emerging technology that enables scientists to measure gene _expression_ across hundreds of thousands of individual cells. These data pose numerous analysis challenges due to the high dimensionality of the data, the numerous interactions
between genes, as well as the technical biases introduced in the experimental process. This week, I will discuss a recent paper that addresses some of these challenges by posing a novel probabilistic generative model of single-cell RNA-seq data. This work
integrates a number of machine learning areas including neural networks, hierarchical Bayesian models, and blackbox variational inference. Furthermore, this work provides a case study for the application of recent machine learning methods to a type of data
that may be unfamiliar to pure machine learning and computer science researchers.