Developing Techniques for the Identification of Non-Canonical RNA Pairing and Analysis of LC-MS Datasets

Authors

Christopher Jurich, University of Nebraska-LincolnFollow

First Advisor

Professor Joseph Yesselman

Second Advisor

Professor Robert Powers

Date of this Version

Winter 12-10-2021

Document Type

Article

Comments

A THESIS Presented to the Faculty of The Graduate College of the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Chemistry, Under the Supervision of Professor Joseph D. Yesselman. Lincoln, Nebraska: December, 2021

Copyright © 2021 Christopher P. Jurich

Abstract

Non-canonical pairing dynamics in ribonucleic acid (RNA) structureand statistical analysis of metabolomics liquid chromatography mass spectrometry (LC-MS) datasets are two difficult problems that stand as open challenges.

RNA folding algorithms are used across a variety of disciplines to predict structures when experimental elucidation techniques are inconvenient or impractical. Though successful and widely adopted, folding algorithms make simplifying assumptions for loop regions due to their complex interactions and associated difficulty with generating energy parameters for relevant non-canonical pairing interactions. Modeling assumptions and a lack of energy parameters for loops limit accuracy in these functional critical regions of RNA. This work describes a new technique for probing non-canonical loop interactions through the combined analysis of dimethyl sulfate (DMS) and three-dimensional crystallographic data. We demonstrate that DMS data encodes information about non-canonical pairing which describes these interactions in an efficient, high throughput manner.

Metabolomics aims to understand biological processes through the analysis of small molecule metabolites. The field primarily uses ¹H nuclear magnetic resonance (NMR) spectroscopy as well as LC-MS to identify and quantitate metabolites. With even simple samples having hundreds or thousands of metabolites, researchers in the field have developed software pipelines to make metabolomics studies a tractable task. Numerous packages exist for the analysis of either ¹H NMR or LC-MS data, but current offerings force researchers to use multiple packages to analyze both data types. To address the need for a metabolomics package capable of analyzing both, we have developed new LC-MS functionality for the NMR metabolomics package MVAPACK.

Advisor: Joseph D. Yesselman