TABLE 2.
Antigenicity associated sites identified in H5N1’s HA protein by using a sparse learning method.
| positiona |
mutationc | weight (SD) using HI datad |
weight (SD) using MN datad |
bootstrap valuee |
Reference | |||
|---|---|---|---|---|---|---|---|---|
| H5 | H3 | H1b | HI | MN | ||||
| 45 | 54(C) | 48 | D/N | 0.163(0.067) | 0.246(0.079) | 65 | 89 | |
| 71 | 80(E) | 76 | I/L | 0.385(0.196) | 0.628(0.195) | 93 | 100 | |
| 83 | 91(E) | 87 | A/I | 0.251(0.166) | 0.200(0.116) | 78 | 78 | |
| 86 | 94(E) | 90 | A/V | 0.412(0.093) | 0.252(0.077) | 100 | 100 | 11; 28 |
| 94 | 101 | 98 | N/D/V | 0.195(0.099) | 0.256(0.090) | 73 | 96 | |
| 120 | 125 | 122 | S/N/D | 0.357(0.110) | 0.231(0.081) | 100 | 98 | |
| 124 | 129(RBS) | 129(Sa) | D/N/S | 0.640(0.165) | 0.165(0.091) | 100 | 81 | 11; 27 |
| 133 | 137(A) | 136 | S/A | 0.192(0.137) | 0.404(0.136) | 74 | 98 | |
| 140 | 144(A) | 144 | K/R/T/G/M/E | 0.249(0.101) | 0.136(0.064) | 91 | 64 | 26 |
| 141 | 145(A) | 145(Ca2) | S/P | 0.169(0.119) | 0.181(0.074) | 64 | 68 | 26 |
| 154 | 158(B) | 158(Sb) | N/D/G | 0.158(0.067) | 0.135(0.052) | 74 | 61 | 28 |
| 156 | 160(B) | 160(Sa) | A/T/K | 0.293(0.094) | 0.155(0.056) | 98 | 92 | 26 |
| 162 | 166 | 166(Sa) | R/I/K/S | 0.321(0.095) | 0.243(0.088) | 94 | 96 | 26 |
| 174 | 178(D) | 178 | V/I | 0.297(0.213) | 0.177(0.125) | 89 | 72 | |
| 175 | 179(D) | 179 | L/M | 0.224(0.113) | 0.399(0.114) | 80 | 91 | |
| 184 | 188(B) | 188 | A/E | 0.222(0.118) | 0.238(0.107) | 81 | 90 | |
| 210 | 214(D) | 214 | V/T/I | 0.210(0.072) | 0.185(0.052) | 82 | 87 | |
| 227 | 231 | 231 | E/D | 0.202(0.128) | 0.216(0.089) | 60 | 84 | |
| 252 | 256 | 256 | Y/N | 0.213(0.116) | 0.186(0.074) | 72 | 72 | |
| 263 | 266 | 267 | A/T | 0.304(0.074) | 0.358(0.056) | 89 | 89 | 27 |
| 282 | 285 | 286 | M/I | 0.305(0.095) | 0.248(0.074) | 89 | 87 | |
a
The position is numbered using the corresponding residue number in the H5 HA sequence or the corresponding one in the H3 HA protein;
b
The position in H1 was based on the amino acid number in HA of PR8 (PDB ID 1RU7);
c
The mutations are predicted to affect the titers in immunologic datasets, and the corresponding antibody binding sites in H3 HA are shown in parentheses. RBS denotes receptor-binding site;
d
The weight was an mean of absolute weight values calculated from a total of 100 runs using HI or MN data, and the numbers in the parenthesis denote the corresponding standard deviations;
e
The bootstrap values were calculated based on 100 runs.