Abstract

In recent studies involving NAPPA microarrays, extra-well fluorescence is used as a key measure for identifying disease biomarkers because there is evidence to support that it is better correlated with strong antibody responses than statistical analysis involving intraspot intensity. Because this feature is not well quantified by traditional image analysis software, identification and quantification of extra-well fluorescence is performed manually, which is both time-consuming and highly susceptible to variation between raters. A system that could automate this task efficiently and effectively would greatly improve the process of data acquisition in microarray studies, thereby accelerating the discovery of disease biomarkers. In this study, we experimented with different machine learning methods, as well as novel heuristics, for identifying spots exhibiting extra-well fluorescence (rings) in microarray images and assigning each ring a grade of 1-5 based on its intensity and morphology. The sensitivity of our final system for identifying rings was found to be 72% at 99% specificity and 98% at 92% specificity. Our system performs this task significantly faster than a human, while maintaining high performance, and therefore represents a valuable tool for microarray image analysis.

Original languageEnglish (US)
Pages (from-to)3969-3977
Number of pages9
JournalJournal of Proteome Research
Volume16
Issue number11
DOIs
StatePublished - Nov 3 2017

Fingerprint

Microarrays
Fluorescence
Biomarkers
Image analysis
Microarray Analysis
Antibody Formation
Heuristic methods
Software
Learning systems
Data acquisition
Statistical methods
Antibodies

Keywords

  • bioinformatics
  • biomarker
  • image analysis
  • nucleic acid programmable protein array (NAPPA)
  • protein array

ASJC Scopus subject areas

  • Biochemistry
  • Chemistry(all)

Cite this

Rivera, R., Wang, J., Yu, X., Demirkan, G., Hopper, M., Bian, X., ... Wallstrom, G. (2017). Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images. Journal of Proteome Research, 16(11), 3969-3977. https://doi.org/10.1021/acs.jproteome.7b00267

Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images. / Rivera, Robert; Wang, Jie; Yu, Xiaobo; Demirkan, Gokhan; Hopper, Marika; Bian, Xiaofang; Tahsin, Tasnia; Magee, Dewey; Qiu, Ji; LaBaer, Joshua; Wallstrom, Garrick.

In: Journal of Proteome Research, Vol. 16, No. 11, 03.11.2017, p. 3969-3977.

Research output: Contribution to journalArticle

Rivera, R, Wang, J, Yu, X, Demirkan, G, Hopper, M, Bian, X, Tahsin, T, Magee, D, Qiu, J, LaBaer, J & Wallstrom, G 2017, 'Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images', Journal of Proteome Research, vol. 16, no. 11, pp. 3969-3977. https://doi.org/10.1021/acs.jproteome.7b00267
Rivera, Robert ; Wang, Jie ; Yu, Xiaobo ; Demirkan, Gokhan ; Hopper, Marika ; Bian, Xiaofang ; Tahsin, Tasnia ; Magee, Dewey ; Qiu, Ji ; LaBaer, Joshua ; Wallstrom, Garrick. / Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images. In: Journal of Proteome Research. 2017 ; Vol. 16, No. 11. pp. 3969-3977.
@article{fdc4c3325664420295f539d50a975523,
title = "Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images",
abstract = "In recent studies involving NAPPA microarrays, extra-well fluorescence is used as a key measure for identifying disease biomarkers because there is evidence to support that it is better correlated with strong antibody responses than statistical analysis involving intraspot intensity. Because this feature is not well quantified by traditional image analysis software, identification and quantification of extra-well fluorescence is performed manually, which is both time-consuming and highly susceptible to variation between raters. A system that could automate this task efficiently and effectively would greatly improve the process of data acquisition in microarray studies, thereby accelerating the discovery of disease biomarkers. In this study, we experimented with different machine learning methods, as well as novel heuristics, for identifying spots exhibiting extra-well fluorescence (rings) in microarray images and assigning each ring a grade of 1-5 based on its intensity and morphology. The sensitivity of our final system for identifying rings was found to be 72{\%} at 99{\%} specificity and 98{\%} at 92{\%} specificity. Our system performs this task significantly faster than a human, while maintaining high performance, and therefore represents a valuable tool for microarray image analysis.",
keywords = "bioinformatics, biomarker, image analysis, nucleic acid programmable protein array (NAPPA), protein array",
author = "Robert Rivera and Jie Wang and Xiaobo Yu and Gokhan Demirkan and Marika Hopper and Xiaofang Bian and Tasnia Tahsin and Dewey Magee and Ji Qiu and Joshua LaBaer and Garrick Wallstrom",
year = "2017",
month = "11",
day = "3",
doi = "10.1021/acs.jproteome.7b00267",
language = "English (US)",
volume = "16",
pages = "3969--3977",
journal = "Journal of Proteome Research",
issn = "1535-3893",
publisher = "American Chemical Society",
number = "11",

}

TY - JOUR

T1 - Automatic Identification and Quantification of Extra-Well Fluorescence in Microarray Images

AU - Rivera, Robert

AU - Wang, Jie

AU - Yu, Xiaobo

AU - Demirkan, Gokhan

AU - Hopper, Marika

AU - Bian, Xiaofang

AU - Tahsin, Tasnia

AU - Magee, Dewey

AU - Qiu, Ji

AU - LaBaer, Joshua

AU - Wallstrom, Garrick

PY - 2017/11/3

Y1 - 2017/11/3

N2 - In recent studies involving NAPPA microarrays, extra-well fluorescence is used as a key measure for identifying disease biomarkers because there is evidence to support that it is better correlated with strong antibody responses than statistical analysis involving intraspot intensity. Because this feature is not well quantified by traditional image analysis software, identification and quantification of extra-well fluorescence is performed manually, which is both time-consuming and highly susceptible to variation between raters. A system that could automate this task efficiently and effectively would greatly improve the process of data acquisition in microarray studies, thereby accelerating the discovery of disease biomarkers. In this study, we experimented with different machine learning methods, as well as novel heuristics, for identifying spots exhibiting extra-well fluorescence (rings) in microarray images and assigning each ring a grade of 1-5 based on its intensity and morphology. The sensitivity of our final system for identifying rings was found to be 72% at 99% specificity and 98% at 92% specificity. Our system performs this task significantly faster than a human, while maintaining high performance, and therefore represents a valuable tool for microarray image analysis.

AB - In recent studies involving NAPPA microarrays, extra-well fluorescence is used as a key measure for identifying disease biomarkers because there is evidence to support that it is better correlated with strong antibody responses than statistical analysis involving intraspot intensity. Because this feature is not well quantified by traditional image analysis software, identification and quantification of extra-well fluorescence is performed manually, which is both time-consuming and highly susceptible to variation between raters. A system that could automate this task efficiently and effectively would greatly improve the process of data acquisition in microarray studies, thereby accelerating the discovery of disease biomarkers. In this study, we experimented with different machine learning methods, as well as novel heuristics, for identifying spots exhibiting extra-well fluorescence (rings) in microarray images and assigning each ring a grade of 1-5 based on its intensity and morphology. The sensitivity of our final system for identifying rings was found to be 72% at 99% specificity and 98% at 92% specificity. Our system performs this task significantly faster than a human, while maintaining high performance, and therefore represents a valuable tool for microarray image analysis.

KW - bioinformatics

KW - biomarker

KW - image analysis

KW - nucleic acid programmable protein array (NAPPA)

KW - protein array

UR - http://www.scopus.com/inward/record.url?scp=85032818945&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032818945&partnerID=8YFLogxK

U2 - 10.1021/acs.jproteome.7b00267

DO - 10.1021/acs.jproteome.7b00267

M3 - Article

VL - 16

SP - 3969

EP - 3977

JO - Journal of Proteome Research

JF - Journal of Proteome Research

SN - 1535-3893

IS - 11

ER -