Title

Comparison of Multiple Methods for Quantifying β-Strand Twist in a Protein Structure

Presenting Author Name/s

Brittany Hasty

Faculty Advisor

Jing He

Faculty Advisor Email

jhe@cs.odu.edu

Presentation Type

Poster

Disciplines

Bioinformatics | Computer Sciences

Description/Abstract

Although proteins have a variety of atomic structures, they are often composed of secondary structures of helices, β-sheets, turns, and loops. β-sheet is a major type of secondary structures and was discovered in the 70’s that all β-sheets are twisted in nature. When images of proteins are produced at low resolutions, it is difficult to distinguish β-strands that are components of a β-sheet. Twist of a β-sheet was previously utilized to predict the position of β-strands in the image of a β-sheet. However, it is challenging to quantify the overall twist precisely. We studied multiple methods in quantifying the twist of a β-sheet. In a test consisting ten beta sheets, comparison shows that the best method varied with the size of the beta sheet. For sheets with three or less strands, overall twist calculated with the average twist angles of the two longest beta strands best fit the true protein structure. In sheets with four or five strands, the average of all angles provided the best fit, and in sheets with six or more strands, the average of all the angles except those on the edges of the beta sheet was the most accurate.

Session Title

Poster Session

Location

Learning Commons @ Perry Library, Northwest Atrium

Start Date

3-2-2018 8:00 AM

End Date

3-2-2018 12:30 PM

This document is currently not available here.

Share

COinS
 
Feb 3rd, 8:00 AM Feb 3rd, 12:30 PM

Comparison of Multiple Methods for Quantifying β-Strand Twist in a Protein Structure

Learning Commons @ Perry Library, Northwest Atrium

Although proteins have a variety of atomic structures, they are often composed of secondary structures of helices, β-sheets, turns, and loops. β-sheet is a major type of secondary structures and was discovered in the 70’s that all β-sheets are twisted in nature. When images of proteins are produced at low resolutions, it is difficult to distinguish β-strands that are components of a β-sheet. Twist of a β-sheet was previously utilized to predict the position of β-strands in the image of a β-sheet. However, it is challenging to quantify the overall twist precisely. We studied multiple methods in quantifying the twist of a β-sheet. In a test consisting ten beta sheets, comparison shows that the best method varied with the size of the beta sheet. For sheets with three or less strands, overall twist calculated with the average twist angles of the two longest beta strands best fit the true protein structure. In sheets with four or five strands, the average of all angles provided the best fit, and in sheets with six or more strands, the average of all the angles except those on the edges of the beta sheet was the most accurate.