Prediction of disulfide bonding sites in proteins using position and composition relative feature vectors

Simple item page
dc.contributor.authorJamil, Mehreen
dc.date.accessioned2018-01-24T06:04:56Z
dc.date.available2018-01-24T06:04:56Z
dc.date.issued2017
dc.descriptionSupervised by: Dr. Yaser Daanial Khanen_US
dc.description.abstractThe presence of disulfide bonds in a protein confers an additional stability to protein against various threats. The formation of correct disulfide bonds between cysteine residues ensures proper folding of the protein during in vivo and in vitro folding process. Oxidation of these bonds may disturb the proper biological activity of a protein. Not all cysteines in a protein are involved in the formation of disulfide bonds,therefore prediction of accurate disulfide bonds is crucial for structural and functional relationship of a protein. Many neurodegenerative diseases are caused by the improper formation of disulfide bonds in the nervous system .The determination of fallacious S-S interaction is crucial for correct diagnosis of these disorders. In this study a novel method is used to predict the intra molecular disulfide bonding accurately using context based information. The surrounding amino acids of the cysteine, involved in disulfide bond, play a vital role in making disulfide bonds and used as feature vectors. The proposed method uses context-based data to calculate statistical moments. Statistical moments are very important as they are very sensitive regarding to position of data sequences. For prediction of intra molecular disulfide bonds, these moments are combined together to train neural networks. 10-fold validation on accumulative dataset gives us 87.52% accurate result. Estimation of accurate disulfide bonding against independent data set for 5-fold and 10-fold result in 82.4% and 88% respectively. The overall accuracy of system is 86.3% to sensitivity value 82.4% and specificity 93%.en_US
dc.identifier.urihttps://escholar.umt.edu.pk/handle/123456789/2568
dc.language.isoenen_US
dc.publisherUniversity of Management and Technologen_US
dc.subjectAdditional stabilityen_US
dc.subjectdisulfide bondsen_US
dc.subjectMCS thesisen_US
dc.titlePrediction of disulfide bonding sites in proteins using position and composition relative feature vectorsen_US
dc.typeThesisen_US
Files
Original bundle
Now showing 1 - 2 of 2
Thumbnail Image
Name:
Summary.pdf
Size:
113.07 KB
Format:
Adobe Portable Document Format
Description:
Download
No Thumbnail Available
Name:
For Full text.htm
Size:
22.2 KB
Format:
Hypertext Markup Language
Description:
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Department of Computer Science