A classification approach for genotyping viral sequences based on multidimensional scaling and linear discriminant analysis

Table 2 The test results with HIV-1 sequences^§

Category	All		Outlierness < 2.0		No. of reference sequences
	Total	% acc.*	Total	% acc.*
(a) by gene segment
Gag	14,915	97.85	14,570	98.65	1,142
pol	64,607	98.67	61,704	99.20	735
vif	1,330	99.62	1,324	100.00	945
vpr	1,713	99.71	1,702	99.71	810
tat	1,987	99.35	1,976	99.44	659
env	63,519	99.17	62,652	99.52	1,175
nef	5,598	99.59	5,572	99.82	1,559
(b) by subtype
A	11,907	95.44	11,317	97.74	259
B	112,997	99.53	111,396	99.64	1,750
C	12,785	99.1	12,586	99.52	908
D	3,743	95.99	3,526	98.24	158
F	1,321	96.14	939	97.12	37
G	2,379	94.33	1,800	95.89	65
H	214	71.03	47	51.06	11
J	124	54.84	15	0	3
K	39	69.23	11	54.55	3
01_AE	8,160	98.80	7,863	98.93	152
Total	153,669	98.86	149,500	99.32	3,346

*% accuracy given as 100 × Matched/Total, where Matched is the number cases concordant between MuLDAS and LANL
^§Major analysis (M-group and CRF01) only.

ISSN: 1471-2105