ComPhy: prokaryotic composite distance phylogenies inferred from whole-genome gene sets

BMC Bioinformatics

Table 2 Accuracy for different components combinations of our proposed distance method

Data sets	Number of species	GDD (%)	GCD (%)	GBD (%)	GCD*GDD (%)	GCD*GBD (%)	GDD*GBD (%)	GCDGDDGBD (%)
Dataset1	52	85.12	86.44	84.54	91.45	90.29	90.29	90.29
Dataset2	53	87.76	86.40	84.45	90.65	90.74	90.74	90.74
Dataset3	82	80.37	92.58	84.19	94.46	95.93	96.06	98.46
Dataset4	398	83.73	86.56	81.23	89.93	87.07	87.28	90.07
Dataset5	181	95.04	89.74	90.20	94.30	95.67	98.16	98.30
Dataset6	96	87.39	85.45	84.88	99.36	99.26	99.36	99.26
Dataset7	277	88.70	84.04	86.75	88.71	89.71	88.23	90.71
Dataset8	165	85.36	77.98	77.03	94.44	94.38	94.47	94.38
Dataset9	54	89.31	87.34	83.76	92.31	92.31	92.37	96.55

GDD = gene dispersion distance, GCD = gene content distance, GBD = gene breakpoint distance. GCD*GDD is the combination of two distance components and GCD*GDD*GBD is the combination of all three terms. All distances are logarithmically transformed.

ISSN: 1471-2105