Dissertation Defense Schedule
Academic Excellence
Sharing original dissertation research is a principle to which the University of Delaware is deeply committed. It is the single most important assignment our graduate students undertake and upon completion is met with great pride.
We invite you to celebrate this milestone by attending their dissertation defense. Please review the upcoming dissertation defense schedule below and join us!
PROGRAM | Bioinformatics & Computational Biology
Machine Learning-Driven Multi-Omics for Optimizing Biotherapeutic Production
By: Sai Guna Ranjan Gurazada Chair: Shawn Polson
ABSTRACT
Biotherapeutics, particularly monoclonal antibodies (mAbs) and recombinant adeno-associated viruses (rAAVs), have transformed the treatment landscape for various diseases. However, their production in mammalian host systems, such as Chinese Hamster Ovary (CHO) and Human Embryonic Kidney (HEK) 293 cells, faces challenges due to uncharacterized molecular mechanisms and gene regulatory processes that impede productivity, product quality, and cell growth. Systems biology and omics technologies offer a powerful approach to elucidate the cellular processes and molecular mechanisms influencing recombinant protein production, however the integration and analysis of large-scale, high-dimensional, multi-omics data generated poses challenges in extracting meaningful biological insights and developing predictive models for optimization.
This dissertation leverages machine learning (ML) approaches to harness the potential of omics data for forecasting and improving recombinant protein production in these mammalian host systems. Chapter 1 outlines the research objectives and motivation. Chapter 2 explores the omics landscape in rAAV-producing HEK cells, identifying gaps and opportunities for leveraging omics to enhance rAAV production. In Chapter 3, a predictive ML model is developed and evaluated using a large-scale transcriptomics dataset from mAb-producing CHO cells, forecasting a critical productivity metric, product titer. Chapter 4 implements a Bayesian ML framework that integrates multi-omics data (transcriptomics, proteomics, and metabolomics) from mAb-producing CHO cells to predict cell culture productivity indicators, including overall mAb potential, and identifies a panel of potential biomarkers indicative of high producers. Finally, Chapter 5 summarizes the key findings and discusses potential future directions, highlighting the potential of ML-driven multi-omics integration to revolutionize bioprocess optimization and biotherapeutic production.
As a result, a multi-omic panel of potential biomarkers was identified, comprising 4 genes, 8 proteins, and 11 metabolites, capable of distinguishing cell cultures based on their mAb production potential. Cross-omics comparisons indicated that proteomics and metabolomics data had a comparatively stronger influence on predicting manufacturing outcomes. Further, an innovative approach was implemented to mitigate the limitations associated with the relatively small sample size of the multi-omics datasets for building predictive ML models. In conclusion, this dissertation demonstrates the potential of ML techniques for extracting valuable insights from complex omics datasets, enabling the development of predictive models and the identification of key biomarkers and cellular pathways for optimizing recombinant protein production in CHO and HEK cell systems.
The Process
Step-by-Step
Visit our “Step-by-Step Graduation Guide” to take you through the graduation process.From formatting your Dissertation to Doctoral Hooding procedures.
Dissertation Manual
Wondering how to set up the format for your paper. Refer to the “UD Thesis/Dissertation Manual” for formatting requirements and more.
Defense Submission Form
This form must be completed two weeks in advance of a dissertation defense to meet the University of Delaware Graduate and Professional Education’s requirements.
