Skip to main content

Table 5 Classification for Parkinson study.

From: MegaSNPHunter: a learning approach to detect disease predisposition SNPs and high level interactions in genome wide association study

Chromosome Picked SNPs Total SNPs Prediction Accuracy Randomized test accuracy
1 242 31,532 0.852 0.505
2 247 32,706 0.874 0.516
3 218 27,691 0.874 0.517
4 174 24,193 0.835 0.511
5 188 24,570 0.878 0.507
6 204 26,372 0.857 0.501
7 278 21,382 0.821 0.498
8 254 22,434 0.845 0.508
9 243 19,542 0.841 0.505
10 227 20,007 0.841 0.507
11 247 19,539 0.854 0.513
12 230 19,572 0.806 0.506
13 156 14,123 0.784 0.502
14 224 12,645 0.824 0.509
15 212 11,618 0.786 0.518
16 225 11,767 0.793 0.496
17 202 11,619 0.778 0.507
18 252 12,613 0.793 0.507
19 165 8,608 0.802 0.5
20 186 10,375 0.806 0.512
21 130 6,612 0.758 0.497
22 126 7,071 0.782 0.506
OVERALL 339 396,588 0.913 0.503
  1. The classification performance of MegaSNPHunter on Parkinson study.