Mason Archival Repository Service

Application of a Topological Descriptor for Protein Interface Identification and Protein Binding Prediction

Show simple item record

dc.contributor.advisor Vaisman, Iosif I.
dc.contributor.author Peters, Olivia
dc.creator Peters, Olivia
dc.date 2010-04-10
dc.date.accessioned 2010-05-11T18:21:42Z
dc.date.available NO_RESTRICTION en_US
dc.date.available 2010-05-11T18:21:42Z
dc.date.issued 2010-05-11T18:21:42Z
dc.identifier.uri http://hdl.handle.net/1920/5805
dc.description.abstract The identification of proteins which interact or form complexes is a critical step in advancing several aspects of computational biology, including intelligent protein design and functional prediction. Previous methods have focused primarily on sequence alignment or threading methods to accomplish this, requiring large libraries of sequences. This work is an attempt to advance the current field of protein prediction through the use of a structural geometry methodology proven successful for many other aspects of proteomic analyses. The method is extended in two ways; first, a classification approach is created to identify protein residues involved in the binding interface, with the intent of using this information to aid the prediction of protein complex formation. Results are promising, with better than eighty percent correct classification, comparable to the best techniques currently in use. Second, a methodology was created to score potential docking conformations. Of the 54 proteins in the test data set, 43 had a near-native structure in the top 100 positions, and a median ratio of successfully identified residue contacts of 0.57. The structural geometry method has been successfully applied to these two problems to advance the state of the field of proteomics. en_US
dc.language.iso en_US en_US
dc.subject proteomics en_US
dc.subject Delaunay en_US
dc.subject interface en_US
dc.subject classification en_US
dc.subject docking en_US
dc.title Application of a Topological Descriptor for Protein Interface Identification and Protein Binding Prediction en_US
dc.type Dissertation en
thesis.degree.name Doctor of Philosophy Bioinformatics and Computational Biology en_US
thesis.degree.level Doctoral en
thesis.degree.discipline Bioinformatics and Computational Biology en
thesis.degree.grantor George Mason University en


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search MARS


Browse

My Account

Statistics