PeakSeg: Constrained Optimal Segmentation and Supervised Penalty Learning for Peak Detection in Count Data

Abstract

Peak detection is a central problem in genomic data analysis, and current algorithms for this task are unsupervised and mostly effective for a single data type and pattern (e.g. H3K4me3 data with a sharp peak pattern). We propose PeakSeg, a new constrained maximum likelihood segmentation model for peak detection with an efficient inference algorithm: constrained dynamic programming. We investigate unsupervised and supervised learning of penalties for the critical model selection problem. We show that the supervised method has state-of-the-art peak detection across all data sets in a benchmark that includes both sharp H3K4me3 and broad H3K36me3 patterns.

Cite

Text

Hocking et al. "PeakSeg: Constrained Optimal Segmentation and Supervised Penalty Learning for Peak Detection in Count Data." International Conference on Machine Learning, 2015.

Markdown

[Hocking et al. "PeakSeg: Constrained Optimal Segmentation and Supervised Penalty Learning for Peak Detection in Count Data." International Conference on Machine Learning, 2015.](https://mlanthology.org/icml/2015/hocking2015icml-peakseg/)

BibTeX

@inproceedings{hocking2015icml-peakseg,
  title     = {{PeakSeg: Constrained Optimal Segmentation and Supervised Penalty Learning for Peak Detection in Count Data}},
  author    = {Hocking, Toby and Rigaill, Guillem and Bourque, Guillaume},
  booktitle = {International Conference on Machine Learning},
  year      = {2015},
  pages     = {324-332},
  volume    = {37},
  url       = {https://mlanthology.org/icml/2015/hocking2015icml-peakseg/}
}