Clustering of Multidimensional Data Sets with Applications to Spatial Distributions of Ribosomal Proteins

Thumbnail Image

Files

REU2013Team3Bio.pdf (427.24 KB)

Links to Files

https://userpages.umbc.edu/~gobbert/papers/REU2013Team3Bio.pdf

Permanent Link

http://hdl.handle.net/11603/11410

Collections

UMBC Mathematics and Statistics Department
UMBC Biological Sciences Department
UMBC Computer Science and Electrical Engineering Department
UMBC Faculty Collection
UMBC Student Collection

Author/Creator

Author/Creator ORCID

Date

2013

Type of Work

techical report
Text

Department

Program

Citation of Original Publication

Rights

This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.

Subjects

ribosomal proteins
Mahalanobis distance
UMBC High Performance Computing Facility (HPCF)
Proteins related to the cofactor phenotype
computation of Mahalanobis distances
determining the optimal pairing

Abstract

Consider ribosomal proteins, each with a three-dimensional spatial location. Proteins related to the cofactor phenotype may be randomly or non-randomly distributed within the ribosome. To investigate this question, the Mahalanobis distance is computed between each pair of protein locations, and the optimal pairing is determined by minimizing the sum of the within-pair distances. Since no single code exists that allows for the computation of Mahalanobis distances, determining the optimal pairing, and determining whether the two groups are statistically different, we created a code that allows a user to do just this. The user can also compute an exact p-value for this distribution rather than rely on an approximation.