CNV detection method optimized for high-resolution arrayCGH by normality test

Jaegyoon Ahn, Youngmi Yoon, Chihyun Park, Sang Hyun Park

Research output: Contribution to journalArticle

Abstract

High-resolution arrayCGH platform makes it possible to detect small gains and losses which previously could not be measured. However, current CNV detection tools fitted to early low-resolution data are not applicable to larger high-resolution data. When CNV detection tools are applied to high-resolution data, they suffer from high false-positives, which increases validation cost. Existing CNV detection tools also require optimal parameter values. In most cases, obtaining these values is a difficult task. This study developed a CNV detection algorithm that is optimized for high-resolution arrayCGH data. This tool operates up to 1500 times faster than existing tools on a high-resolution arrayCGH of whole human chromosomes which has 42 million probes whose average length is 50 bases, while preserving false positive/negative rates. The algorithm also uses a normality test, thereby removing the need for optimal parameters. To our knowledge, this is the first formulation for CNV detecting problems that results in a near-linear empirical overall complexity for real high-resolution data.

Original languageEnglish
Pages (from-to)468-473
Number of pages6
JournalComputers in Biology and Medicine
Volume42
Issue number4
DOIs
Publication statusPublished - 2012 Apr 1

Fingerprint

Human Chromosomes
Costs and Cost Analysis
Chromosomes
Costs

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Health Informatics

Cite this

Ahn, Jaegyoon ; Yoon, Youngmi ; Park, Chihyun ; Park, Sang Hyun. / CNV detection method optimized for high-resolution arrayCGH by normality test. In: Computers in Biology and Medicine. 2012 ; Vol. 42, No. 4. pp. 468-473.
@article{e1a10614ebcb4463a90cfb7984f684b5,
title = "CNV detection method optimized for high-resolution arrayCGH by normality test",
abstract = "High-resolution arrayCGH platform makes it possible to detect small gains and losses which previously could not be measured. However, current CNV detection tools fitted to early low-resolution data are not applicable to larger high-resolution data. When CNV detection tools are applied to high-resolution data, they suffer from high false-positives, which increases validation cost. Existing CNV detection tools also require optimal parameter values. In most cases, obtaining these values is a difficult task. This study developed a CNV detection algorithm that is optimized for high-resolution arrayCGH data. This tool operates up to 1500 times faster than existing tools on a high-resolution arrayCGH of whole human chromosomes which has 42 million probes whose average length is 50 bases, while preserving false positive/negative rates. The algorithm also uses a normality test, thereby removing the need for optimal parameters. To our knowledge, this is the first formulation for CNV detecting problems that results in a near-linear empirical overall complexity for real high-resolution data.",
author = "Jaegyoon Ahn and Youngmi Yoon and Chihyun Park and Park, {Sang Hyun}",
year = "2012",
month = "4",
day = "1",
doi = "10.1016/j.compbiomed.2011.12.015",
language = "English",
volume = "42",
pages = "468--473",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier Limited",
number = "4",

}

CNV detection method optimized for high-resolution arrayCGH by normality test. / Ahn, Jaegyoon; Yoon, Youngmi; Park, Chihyun; Park, Sang Hyun.

In: Computers in Biology and Medicine, Vol. 42, No. 4, 01.04.2012, p. 468-473.

Research output: Contribution to journalArticle

TY - JOUR

T1 - CNV detection method optimized for high-resolution arrayCGH by normality test

AU - Ahn, Jaegyoon

AU - Yoon, Youngmi

AU - Park, Chihyun

AU - Park, Sang Hyun

PY - 2012/4/1

Y1 - 2012/4/1

N2 - High-resolution arrayCGH platform makes it possible to detect small gains and losses which previously could not be measured. However, current CNV detection tools fitted to early low-resolution data are not applicable to larger high-resolution data. When CNV detection tools are applied to high-resolution data, they suffer from high false-positives, which increases validation cost. Existing CNV detection tools also require optimal parameter values. In most cases, obtaining these values is a difficult task. This study developed a CNV detection algorithm that is optimized for high-resolution arrayCGH data. This tool operates up to 1500 times faster than existing tools on a high-resolution arrayCGH of whole human chromosomes which has 42 million probes whose average length is 50 bases, while preserving false positive/negative rates. The algorithm also uses a normality test, thereby removing the need for optimal parameters. To our knowledge, this is the first formulation for CNV detecting problems that results in a near-linear empirical overall complexity for real high-resolution data.

AB - High-resolution arrayCGH platform makes it possible to detect small gains and losses which previously could not be measured. However, current CNV detection tools fitted to early low-resolution data are not applicable to larger high-resolution data. When CNV detection tools are applied to high-resolution data, they suffer from high false-positives, which increases validation cost. Existing CNV detection tools also require optimal parameter values. In most cases, obtaining these values is a difficult task. This study developed a CNV detection algorithm that is optimized for high-resolution arrayCGH data. This tool operates up to 1500 times faster than existing tools on a high-resolution arrayCGH of whole human chromosomes which has 42 million probes whose average length is 50 bases, while preserving false positive/negative rates. The algorithm also uses a normality test, thereby removing the need for optimal parameters. To our knowledge, this is the first formulation for CNV detecting problems that results in a near-linear empirical overall complexity for real high-resolution data.

UR - http://www.scopus.com/inward/record.url?scp=84862826319&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862826319&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2011.12.015

DO - 10.1016/j.compbiomed.2011.12.015

M3 - Article

C2 - 22270227

AN - SCOPUS:84862826319

VL - 42

SP - 468

EP - 473

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

IS - 4

ER -