Learning to discover faulty spots in cDNA microarrays

Authors:
Mónica G. Larese;Pablo M. Granitto;Juan C. Gómez
Affiliations:
French Argentine International Center for Information and Systems Sciences, UPCAM, France and UNR, CONICET, Rosario, Argentina and Lab. for System Dynamics and Signal Proc., FCEIA, Univ. Nacional ...;French Argentine International Center for Information and Systems Sciences, UPCAM, France and UNR, CONICET, Rosario, Argentina;French Argentine International Center for Information and Systems Sciences, UPCAM, France and UNR, CONICET, Rosario, Argentina and Lab. for System Dynamics and Signal Proc., FCEIA, Univ. Nacional ...
Venue:
IBERAMIA'10 Proceedings of the 12th Ibero-American conference on Advances in artificial intelligence
Year:
2010

Citing 10
Cited 0

A theory of the learnable

Communications of the ACM
The Strength of Weak Learnability

Machine Learning
The nature of statistical learning theory

The nature of statistical learning theory
A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Image Analysis for Detecting Faulty Spots from Microarray Images

DS '02 Proceedings of the 5th International Conference on Discovery Science
A supervised data-driven approach for microarray spot quality classification

Pattern Analysis & Applications
Boosting Sex Identification Performance

International Journal of Computer Vision
Segmentation of cDNA microarray images by kernel density estimation

Journal of Biomedical Informatics
Logistic ensembles of Random Spherical Linear Oracles for microarray classification

International Journal of Data Mining and Bioinformatics
Boosting image classification with LDA-based feature combination for digital photograph management

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Gene expression ratios obtained from microarray images are strongly affected by the algorithms used to process them as well as by the quality of the images. Hundreds of spots often suffer from quality problems caused by the manufacturing process and many must be discarded because of lack of reliability. Recently, several computational models have been proposed in the literature to identify defective spots, including the powerful Support Vector Machines (SVMs). In this paper we propose to use different strategies based on aggregation methods to classify the spots according to their quality. On one hand we apply an ensemble of classifiers, in particular three boosting methods, namely Discrete, Real and Gentle AdaBoost. As we use a public dataset which includes the subjective labeling criteria of three human experts, we also evaluate different ways of modeling consensus between the experts. We show that for this problem ensembles achieve improved classification accuracies over alternative state-of-the-art methods.