Loading Events

« All Events

  • This event has passed.

Ph.D. Defense: Yue Zhao

November 25, 2019 @ 2:00 pm - 3:00 pm UTC-5

Doctoral Dissertation Defense

Title: Deep Pathway Analysis

Ph.D. Candidate: Yue Zhao

Major Advisor: Dr. Dong-Guk Shin

Associate Advisors: Dr. Lynn Kuo, Dr. Jinbo Bi, Dr. Sheida Nabavi, Dr. Charles Giardina

Day/Time: Monday, November 25th, 2019 2:00 PM

Location: HBL 1947 Conference Room

Abstract:

In this era of biomedical big data, scientists produce genomics data of multiple types (e.g., microarray, RNA-seq, DNA-seq, ChIP-seq, proteomics data, etc. ) and compare it with the prior known gene regulation relationships which are typically organized into curated molecular pathways. This pathway driven multi-omics data analysis helps scientists gain more accurate and interpretable results. However, this pathway analysis research is still at its infancy.

 

The dissertation presents four different ways to improve pathway analysis: (i) Deep pathway analysis utilizing mutation and expression data – it proposes a new way of analyzing biological pathways using Bayesian network model in which the analysis combines both transcriptome data and mutation information and uses the outcome to identify “routes” of aberrant pathways potentially responsible for the etiology of disease; (ii) Deep pathway analysis in algorithmic form and exploiting hyperparameter tuning – It extends the first approach so that the given gene expression data can recognize which portion of sub-pathways are actively utilized in the biological system being studied without pre-defining pathway routes; (iii) Deep pathway analysis incorporating multi-omics data – The established method is further extended to combine multiple heterogeneous omics data types in its analysis to improve prediction accuracy; and (iv) Deep pathway builder using recurrent neural network –  The developed pathway analysis method is used to train a recurrent neural network (RNN) which can be used to construct gene regulatory networks, thus to extend the existing pathways.

 

In summary, we present a series of methods which aim to uncover pathway routes (as opposed to the whole pathway) as the unit of analysis to pinpoint perturbed signals from various omics data sets. we are optimistic that our methods can be applied to studying a wide array of biological systems. The availability of our methods can encourage wet lab scientists to generate multiple types of omics data sets that can be combined to derive improved interpretation of biological systems.

 

Details

Date:
November 25, 2019
Time:
2:00 pm - 3:00 pm UTC-5
Event Category:

Venue

HBL Class of 1947 Conference Room
UConn Library, 369 Fairfield Way, Unit 1005
Storrs, CT 06269 United States
+ Google Map
Phone:
(860) 486-2518
Website:
https://lib.uconn.edu/

Connect With Us