Dissertation/Thesis Abstract

Predicting CRISPR/Cas9 Negative Selection Outcomes from sgRNA Sequences
by Stanton, TImothy, Master's, Stevens Institute of Technology, 2019, 40; 27669807
Abstract (Summary)

Although CRISPR/Cas9 has been said to be a 'reasonably specific' method of gene editing, which performs double strand breaks in DNA to remove or add nucleotides, there have been many cases of large scale deletions of nucleotides that were not targeted and 'complex genomic rearrangements' (Kosicki 2018). Given these unwanted off-target deletions that occur under CRISPR/Cas9 gene editing, a statistical method of analysis was performed to predict, given the sgRNA target specific sequence, whether the CRISPR/Cas9 negative selection screening would be successful and result in the removal of one or two nucleotides from the target strand resulting in a frameshift and removing a single protein or isoform. The final findings of the logistic model are surprisingly definitive. First, we notice that the thymine and guanine nucleotides make the most difference in determining the success of the Cas9 screening. Second, with even greater importance, upon looking at the 5’–17th nucleotide position in the sgRNA sequence, which is the specific locus of Cas9 mediated cleavage, we discover that there is a much greater chance of success if the sgRNA nucleotide is a guanine, and a much greater chance of failure if it is a thymine. Similarly, a guanine in the 5’–19th position will lead to a much greater chance of successful cleavage.

Indexing (document details)
Advisor: Florescu, Ionut
Commitee: Calhoun, George
School: Stevens Institute of Technology
Department: Financial Engineering
School Location: United States -- New Jersey
Source: MAI 81/7(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Genetics, Artificial intelligence, Biostatistics
Keywords: Cas9, CRISPR, Frameshift, Genetics, Prediction, sgRNA
Publication Number: 27669807
ISBN: 9781392569306
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest