Reliable Change Point Detection for ACGH data

Charalambos Eliades, Harris Papadopoulos
Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 230:387-405, 2024.

Abstract

This study introduces two algorithms based on the Inductive Conformal Martingale (ICM) approach to address the change point (CP) detection problem in array-based Comparative Genomic Hybridization (aCGH) data. The ICM, a distribution-free approach with minimal assumptions, is particularly suitable for this application. We have implemented two ICM-based algorithms; the first utilizes nonconformities from preprocessed data, while the second incorporates the label conditional distribution and the labels’ distribution to enhance detection accuracy. This approach significantly improves our results, demonstrating the potential of ICM in complex genomic data analysis.

Cite this Paper

BibTeX
@InProceedings{pmlr-v230-eliades24a, title = {Reliable Change Point Detection for ACGH data}, author = {Eliades, Charalambos and Papadopoulos, Harris}, booktitle = {Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {387--405}, year = {2024}, editor = {Vantini, Simone and Fontana, Matteo and Solari, Aldo and Boström, Henrik and Carlsson, Lars}, volume = {230}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v230/main/assets/eliades24a/eliades24a.pdf}, url = {https://proceedings.mlr.press/v230/eliades24a.html}, abstract = {This study introduces two algorithms based on the Inductive Conformal Martingale (ICM) approach to address the change point (CP) detection problem in array-based Comparative Genomic Hybridization (aCGH) data. The ICM, a distribution-free approach with minimal assumptions, is particularly suitable for this application. We have implemented two ICM-based algorithms; the first utilizes nonconformities from preprocessed data, while the second incorporates the label conditional distribution and the labels’ distribution to enhance detection accuracy. This approach significantly improves our results, demonstrating the potential of ICM in complex genomic data analysis.} }
Endnote
%0 Conference Paper %T Reliable Change Point Detection for ACGH data %A Charalambos Eliades %A Harris Papadopoulos %B Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2024 %E Simone Vantini %E Matteo Fontana %E Aldo Solari %E Henrik Boström %E Lars Carlsson %F pmlr-v230-eliades24a %I PMLR %P 387--405 %U https://proceedings.mlr.press/v230/eliades24a.html %V 230 %X This study introduces two algorithms based on the Inductive Conformal Martingale (ICM) approach to address the change point (CP) detection problem in array-based Comparative Genomic Hybridization (aCGH) data. The ICM, a distribution-free approach with minimal assumptions, is particularly suitable for this application. We have implemented two ICM-based algorithms; the first utilizes nonconformities from preprocessed data, while the second incorporates the label conditional distribution and the labels’ distribution to enhance detection accuracy. This approach significantly improves our results, demonstrating the potential of ICM in complex genomic data analysis.
APA
Eliades, C. & Papadopoulos, H.. (2024). Reliable Change Point Detection for ACGH data. Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 230:387-405 Available from https://proceedings.mlr.press/v230/eliades24a.html.

Related Material