A fully interpretable approach to sequence-to-expression modeling

Bhogale, Shounak Girish

Permalink

https://hdl.handle.net/2142/120529 Copy

Description

Title

A fully interpretable approach to sequence-to-expression modeling

Author(s)

Bhogale, Shounak Girish

Issue Date

2023-04-26

Director of Research (if dissertation) or Advisor (if thesis)

Sinha, Saurabh

Doctoral Committee Chair(s)

Sinha, Saurabh

Committee Member(s)

• Zhao, Dave
• Shukla, Diwakar
• Maslov, Sergei

Department of Study

School of Molecular & Cell Bio

Discipline

Biophysics & Quant Biology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• GRN
• Sequence-to-Expression Modeling

Abstract

Gene expression regulation is a complex process where many regulatory elements such as transcription factors, enhancers, promoters, histone modifications work in tandem. In this thesis we focus our attention primarily on the interplay between two of the most important elements – transcription factors and enhancers. The binding of TFs to enhancers plays crucial role in modulating the gene expression. First, we present a thermodynamics-based sequence-to- expression model to study Drosophila mesodermal differentiation program. This model suggests indirect interaction between a transcription factor and DNA as a plausible mechanism for a transcription factor having dual regulatory roles (as an activator and a repressor) based on its environment. In the second part we discuss a neural network-based sequence-to-expression model to study neuronal stem cell differentiation program. The model learns the effect of transcription factors on gene expression across multiple cell types. It can also be used as a tool for de novo motif discovery or to improve the known motifs based on new data.

Graduation Semester

2023-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Shounak Bhogale

Owning Collections