A Latent Variable Model for Chemogenomic Profiling
Patrick Flaherty, Guri Giaever, Jochen Kumm, Michael I. Jordan, Adam P. Arkin
This is the derivation of the Labeled Latent Dirichlet Allocation Model.
The variable names correspond to those described in the Methods section of the paper.
This hierarchical clustering plot was generated in R.
The dendrogram was generated using
agglomerative clustering on the correlation distance metric with average linkage clustering.
The code for llda is contained in this zip file.[zip]
- Right Click to download these .mat files for matlab.
- The fitness defect data used for both hierarchical clustering and LLDA is in the
"fdscore" data structure and the "wordcount" matrix.
Each row is an ORF and each
column is an experiment. Fitness Defect Data [mat]
Gene Inforamtion [mat]
Treatment Information [mat]