Table 2 The number of markers required in each method for assigning an animal to its breed at different threshold levels of the LLR analysis.

From: Detecting SNP markers discriminating horse breeds by deep learning

Log(10)

90%

95%

98%

DNN

Garson

Olden

DNN

Garson

Olden

DNN

Garson

Olden

1

144

264

205

202

298

242

348

358

363

2

179

291

236

287

327

293

386

419

394

3

230

315

264

330

368

342

*

*

*

4

276

357

287

378

427

380

*

*

*

  1. To calculate these numbers, the sophisticated analysis of LLR was performed individually with a certain number of markers. Using 460 top-rank SNP markers selected by each method, in each step of the LLR analysis, one SNP marker was added to the rest of the markers used in the earlier step.
  2. *It requires more than 460 markers.
Back to article page