Effects of using coding potential, sequence conservation and mRNA structure conservation for predicting pyrrolysine containing genes

Table 2 Cluster ranking

	mttB	mtbB	mtmB	transposase₁	tranposase₂	TetR	p-value
f ^organisms	11	9	10	2	1	2	-
f ^diversity	32.2	18.8	25.2	5.4	0.0	18.0	-
f ^structure	19.0	37.6	22.5	41.3	42.4	23.2	-
f ^coding	111.8	107.1	95.2	103.2	56.0	58.3	-
f ^upstream	1016.6	992.1	605.4	293.2	325.7	297.5	-
f ^downstream	495	325.4	773.2	1150	1149.7	391	-
f ^syn_codons	90.6	96	93.7	98.4	100	91.0	-
f ^size	18	18	19	16	3	2	-
rank(f^organisms)	1	3	2	84	508	85	0.0006
rank(f^diversity)	32	222	87	639	890	243	0.13
rank(f^structure)	895	344	858	196	147	840	0.72
rank(f^coding)	22	26	38	31	80	76	0.00006
rank(f^upstream)	6	7	13	46	32	41	0.000027
rank(f^downstream)	6	90	18	602	888	255	0.064
rank(f^syn_codons)	793	385	596	252	99	762	0.51
rank(f^size)	8	9	6	10	305	498	0.0013
$ranks (\sqrt{{\hat{f}}^{coding} \times {\hat{f}}^{upstream}})$	4	5	10	16	18	19	0.000017
$rank (\sqrt[3]{{\hat{f}}^{structure} \times {\hat{f}}^{diversity} \times {\hat{f}}^{syn_codons}})$	211	6	132	504	876	417	0.13
rank(regression)	2	3	4	15	23	19	0.000015

Raw feature values and ranking of known true positive clusters based on single features and combined ranking using regression weights. p-values for significant features (p > 0:05) are marked in bold.

ISSN: 1471-2105