Applications of Deep-Learning in Genomics Research
University of Michigan School of Public Health
1690 SPH I, 1415 Washington Heights Ann Arbor, MI 48109-2029
Genomics data, and particularly multi-omics genomics data, generally have larger feature sizes than the sample sizes, posing challenges for deep-learning application in this field. In this talk, I will elaborate how we get around the curse of small population size, and apply deep-learning creatively to predict disease prognosis at the population level. We have developed a tool called Cox-nnet that uses gene expression data to predict patients survival via neural network. We further developed another integration tool called DeepProg, which uses multiple types of genomics data to predict patients survival via autoencoders. We demonstrate the utility of these methods on tens of thousands of cancer samples in the cancer genome atlas (TCGA). Lastly, I will present our computational method, called DeepImpute, that uses deep-learning to impute the noisy single-cell RNA-Seq data and achieves high accuracy with efficiency and scalability. Light refreshments for seminar guests will be served at 3:10 p.m. Department of Biostatistics

Applications of Deep-Learning in Genomics Research

Lana Garmire, PhD, Associate Professor - Department of Computational Medicine & Bioinformatics - University of Michigan

icon to add this event to your google calendarApril 25, 2019
3:30 pm - 5:00 pm
1690 SPH I
1415 Washington Heights
Ann Arbor, MI 48109-2029
Sponsored by: Department of Biostatistics
Contact Information: Zhenke Wu (zhenkewu@umich.edu) & Peisong Han (peisong@umich.edu)

Genomics data, and particularly multi-omics genomics data, generally have larger feature sizes than the sample sizes, posing challenges for deep-learning application in this field. In this talk, I will elaborate how we get around the curse of small population size, and apply deep-learning creatively to predict disease prognosis at the population level. We have developed a tool called Cox-nnet that uses gene expression data to predict patients survival via neural network. We further developed another integration tool called DeepProg, which uses multiple types of genomics data to predict patients survival via autoencoders. We demonstrate the utility of these methods on tens of thousands of cancer samples in the cancer genome atlas (TCGA). Lastly, I will present our computational method, called DeepImpute, that uses deep-learning to impute the noisy single-cell RNA-Seq data and achieves high accuracy with efficiency and scalability. Light refreshments for seminar guests will be served at 3:10 p.m.