Table 1.

The BHI scores for the galactose utilization dataset

Similarity matrix	w	No. of genes	BHI (all)	BHI (bp)	BHI (mf)	BHI (cc)
Fused genes	0.5	51	0.49 ± 0.03	0.43 ± 0.05	0.40 ± 0.04	0.49 ± 0.03
Fused genes	1	205	0.37 ± 0.01	0.22 ± 0.01	0.19 ± 0.01	0.37 ± 0.01
Unfused (expression only)	0	205	0.38 ± 0.01	0.26 ± 0.02	0.22 ± 0.02	0.38 ± 0.01
Unfused (expression only)	0.5	154	0.37 ± 0.03	0.30 ± 0.02	0.23 ± 0.02	0.37 ± 0.03
Unfused (ChIP chip only)	0	205	0.28 ± 0.03	0.13 ± 0.01	0.11 ± 0.02	0.25 ± 0.03
Unfused (ChIP chip only)	0.5	154	0.20 ± 0.06	0.06 ± 0.03	0.07 ± 0.04	0.19 ± 0.07
Context-averaged (Liu et al.)	0	205	0.38 ± 0.01	0.26 ± 0.02	0.22 ± 0.01	0.38 ± 0.01
Context-averaged	0.5	205	0.40 ± 0.01	0.24 ± 0.01	0.20 ± 0.01	0.40 ± 0.02
Context-averaged	1	205	0.37 ± 0.01	0.22 ± 0.01	0.19 ± 0.01	0.37 ± 0.01

We compute the BHI scores for each GO (biological process, molecular function and cellular component) and an overall value. The fused genes are those with a posterior probability of being fused of at least 0.5. All other genes are classed as unfused. Context-averaged similarity matrices are simply constructed by averaging the posterior similarity matrix over both contexts (i.e. datasets). This is the method used by Liu et al. For comparison, the result obtained using the BHC algorithm on the gene expression data alone is 0.323.