國立高雄大學圖資館 |

Language: English

Back

Geometric optimization in some proxi...

Panigrahi, Satish Chandra.

Geometric optimization in some proximity and bioinformatics problems.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Geometric optimization in some proximity and bioinformatics problems.
Author:	Panigrahi, Satish Chandra.
Description:	218 p.
Notes:	Source: Dissertation Abstracts International, Volume: 76-03(E), Section: B.
Notes:	Adviser: Asish Mukhopadhyay.
Contained By:	Dissertation Abstracts International76-03B(E).
Subject:	Computer science.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3645306
ISBN:	9781321341447

Geometric optimization in some proximity and bioinformatics problems.
Panigrahi, Satish Chandra.

Geometric optimization in some proximity and bioinformatics problems. - 218 p.

Source: Dissertation Abstracts International, Volume: 76-03(E), Section: B.

Thesis (Ph.D.)--University of Windsor (Canada), 2014.

This item must not be sold to any third party vendors.

The theme of this dissertation is geometric optimization and its applications. We study geometric proximity problems and several bioinformatics problems with a geometric content, requiring the use of geometric optimization tools.

ISBN: 9781321341447Subjects--Topical Terms:

199325
Computer science.

Geometric optimization in some proximity and bioinformatics problems.
LDR:03362nmm a2200337 4500 001 457736
005 20150805065229.5
008 150916s2014 ||||||||||||||||| ||eng d
020 $a 9781321341447
035 $a (MiAaPQ)AAI3645306
035 $a AAI3645306
040 $a MiAaPQ $c MiAaPQ
100 1 $a Panigrahi, Satish Chandra. $3 708800
245 1 0 $a Geometric optimization in some proximity and bioinformatics problems.
300 $a 218 p.
500 $a Source: Dissertation Abstracts International, Volume: 76-03(E), Section: B.
500 $a Adviser: Asish Mukhopadhyay.
502 $a Thesis (Ph.D.)--University of Windsor (Canada), 2014.
506 $a This item must not be sold to any third party vendors.
506 $a This item must not be added to any third party search indexes.
520 $a The theme of this dissertation is geometric optimization and its applications. We study geometric proximity problems and several bioinformatics problems with a geometric content, requiring the use of geometric optimization tools.
520 $a We have investigated the following type of proximity problems. Given a point set in a plane with n distinct points, for each point in the set find a pair of points from the remaining points in the set such that the three points either maximize or minimize some geometric measure defined on these. The measures include (a) sum and product; (b) difference; (c) line-distance; (d) triangle area; (e) triangle perimeter; (f) circumcircle-radius; and (g) triangle-distance in three dimensions.
520 $a We have also studied the application of a linear time incremental geometric algorithm to test the linear separability of a set of blue points from a set of red points, in two and three-dimensional Euclidean spaces. We have used this geometric separability tool on 4 different gene expression data-sets, enumerating gene-pairs and gene-triplets that are linearly separable. Pushing on further, we have exploited this novel tool to identify some bio-marker genes for a classifier. The gene selection method proposed in the dissertation exhibits good classification accuracy as compared to other known feature (or gene) selection methods such as t-values, FCS (Fisher Criterion Score) and SAM (Significance Analysis of Microarrays). Continuing this line of investigation further, we have also designed an efficient algorithm to find the minimum number of outliers when the red and blue point sets are not fully linearly separable.
520 $a We have also explored the applicability of geometric optimization techniques to the problem of protein structure similarity. We have come up with two new algorithms, EDAlignres and EDAlign sse, for pairwise protein structure alignment. EDAlign res identifies the best structural alignment of two equal length proteins by refining the correspondence obtained from eigendecomposition and to maximize the similarity measure for the refined correspondence. EDAlignsse, on the other hand, does not require the input proteins to be of equal length. These have been fully implemented and tested against well-established protein alignment programs.
590 $a School code: 0115.
650 4 $a Computer science. $3 199325
650 4 $a Bioinformatics. $3 194415
690 $a 0984
690 $a 0715
710 2 $a University of Windsor (Canada). $b Computer Science. $3 708801
773 0 $t Dissertation Abstracts International $g 76-03B(E).
790 $a 0115
791 $a Ph.D.
792 $a 2014
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3645306