Differential stepwise evolution of SARS coronavirus functional proteins in different host species

Table 3 REL analysis results for three datasets

Groups^a	No. of sequences^d	Mean d_N-d_S^e	Positively selected site(s)
Spike
BAT	4	-0.957
PC03	3	-0.904
HPEM	14	0.061	49, 75, 77, 144, 239, 244, 311, 344, 778, 860, 861, 1001, 1148, 1163, 1179, 1247
HPL	11	-0.138
PCHP04	27	0.938
SARS^b	40	0.361	75, 147, 227, 239, 243, 244, 311, 462, 479, 609, 613, 743, 765, 778, 1080, 1163
Replicase
BAT	4	-0.985
HP03	21	0.008	654
PCHP04	8	-0.774
SARS	31	-0.561
3'-end ORFs
BAT	4	-0.91
HPEM	5	-0.42
HPML	33	0.152
PCHP04	12	-0.571
SARS^c	40	-0.301

a. At least 3 sequences are needed for REL analysis, so PC03 groups of dataset 2 and 3 were not analyzed.
b. The upper limit in number of sequences for REL test is 40, so 15 sequences were removed from original SARS group (removed sequences' number: 33, 40, 43, 50, 86, 108, 110, 111, 123, 135, 139, 144, 147, 152, 156)
c. 12 sequences were removed from original SARS group (removed sequences' number: 5, 19, 50, 56, 57, 81, 82, 88, 91, 103, 107, 111)
d. As a rule of thumb, at least 10 sequences are needed to detect selection at a signal site with reliability. So some of the results may be not reliable because of not enough sequences are available for some groups.
e. Because d_Scould be 0 for some sites, Datamonkey reports d_N-d_Sin place of d_N/d_S

ISSN: 2730-7182