Skip to main content

Table 1 Prediction performance in simulated public dataset

From: KAML: improving genomic prediction accuracy of complex traits using machine learning determined parameters

Traits/CORa (time in h)b

Methods

LMM

BSLMM

BayesR

KAML

T1

0.732 (0.002)

0.791 (3.875)

0.795 (0.782)

0.801 (0.038)

T2

0.771 (0.002)

0.831 (4.483)

0.832 (0.714)

0.843 (0.043)

T3

0.758 (0.002)

0.827 (4.721)

0.832 (0.740)

0.832 (0.039)

Average

0.754

0.816

0.820

0.825

  1. CORa: The Pearson correlation coefficient between predicted values and additive genetic effects values. The reference dataset included 3000 individuals and 1000 validation individuals
  2. (Time in h)b: The computing time is recorded in hours