Spectral Learning of Large Structured HMMs for Comparative Epigenomics

Authors: Chicheng Zhang, Jimin Song, Kamalika Chaudhuri, Kevin Chen

NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We provide sample complexity bounds for our algorithm and evaluate it experimentally on biological data from nine human cell types. Finally we implement our algorithm and evaluate it on biological data from nine human cell types [7].
Researcher Affiliation Academia Chicheng Zhang UC San Diego chz038@eng.ucsd.edu Jimin Song Rutgers University song@dls.rutgers.edu Kevin C Chen Rutgers University kcchen@dls.rutgers.edu Kamalika Chaudhuri UC San Diego kamalika@eng.ucsd.edu
Pseudocode Yes Algorithm 1 shows how to recover the observation matrices Ou at each node u. Once the Ous are recovered, one can use standard techniques to recover T and W; details are described in Algorithm 2 in the Appendix. ... (See Algorithm 3 in the Appendix for details).
Open Source Code No The paper does not provide any statement about open-sourcing the code or a link to a code repository.
Open Datasets Yes We ran our algorithm, which we call Spectacle-Tree , on a chromatin dataset on human chromosome 1 from nine cell types (H1-h ESC, GM12878, Hep G2, HMEC, HSMM, HUVEC, K562, NHEK, NHLF) from the ENCODE project [7].
Dataset Splits No The paper describes the dataset used and evaluation against CAGE data for promoter prediction, but it does not specify explicit training, validation, and test dataset splits with percentages or sample counts.
Hardware Specification No The paper mentions that their Python implementation ran out of memory and that Spectacle-Tree was implemented in Matlab, but it does not specify any hardware details like CPU/GPU models, memory, or specific computing environments.
Software Dependencies No The paper mentions
Experiment Setup Yes We set the number of hidden states, which we interpret as chromatin states, to m = 6, similar to the choice of ENCODE.