Author 1
Name: Robin Dowell
Org: University of Colorado
Country: United States
Email: robin.dowell@colorado.edu
Author 2
Name: Timothy Danford
Org: MIT
Country: United States
Email: tdanford@gmail.com

Author 3
Name: David Gifford
Org: MIT
Country: United States
Email: dkg@mit.edu

Machine learning techniques such as segmentation, labeling, and clustering have proven invaluable in computational biology. I will describe STEREO, an algorithm for discovering cis-regulatory RNA interactions by assemblying complete and potentially overlapping same-strand RNA transcripts from tiling expression data. STEREO first 
identifies coherent segments of transcription using a segmentation algorithm and then discovers individual transcripts consistent with the observed segments given the intensity and shape constraints by assuming a mixture model. We used STEREO to identify 1446 regions 
of overlapping transcription in two strains of yeast, including transcripts that comprise a new form of molecular toggle switch that controls gene variegation.