Supporting materials for the paper 'Discovering Transcriptional Modules by Bayesian Data Integration'


Here are the PDFs for the figures in the paper.

GO annotation matrix (galactose utilisation data)

ChIP data, sorted by cluster order (galactose utilisation data)

Posterior similarity matrix (galactose utilisation data)

Plots of BHI score versus number of fused genes (galactose utilisation data)

Some 1D marginal posterior distributions (galactose utilisation data)

________________________________

GO annotation matrix (cell cycle data)

ChIP data, sorted by cluster order (cell cycle data)

Posterior similarity matrix (cell cycle data)

Some 1D marginal posterior distributions (cell cycle data)

________________________________

We also generate .csv files containing the results of our analyses looking for over-represented Gene Ontology (GO) terms in each cluster.

Cluster membership (galactose utilisation data)

GO analysis (galactose utilisation data)

Cluster membership (cell cycle data)

GO analysis (cell cycle data)

________________________________

Here are the equivalent results for the unfused genes, clustered solely on the basis of gene expression data.

(unfused) GO annotation matrix (galactose utilisation data)

(unfused) ChIP data, sorted by cluster order (galactose utilisation data)

(unfused) Posterior similarity matrix (galactose utilisation data)

(unfused) Some 1D marginal posterior distributions (galactose utilisation data)

(Some of the plot labels for the unfused cell cycle data aren't very legible, for which we apologise. This was due to a problem with R that we weren't able to resolve)

And the GO analysis for the unfused genes.

(unfused) Cluster membership (galactose utilisation data)

(unfused) GO analysis (galactose utilisation data)


Galactose analysis using the Harbison ChIP data

GO Annotation Matrix (galactose + Harbison data)

Plots of BHI score versus number of fused genes_(galactose+Harbison data)

Gene clustering partition (galactose + Harbison data)

GO analysis_(galactose + Harbison data)

1D marginal posterior distributions_(galactose + Harbison data)

Posterior similarity matrix_(galactose + Harbison data)

ChIP data, sorted by cluster order_(galactose + Harbison data)