Table 3.
Mean computational time of the seven methods in the model fitting stage for 12 traits across three data sets
| Data | Traits | BVSR | rjMCMC | BayesR | LMM | MultiBLUP | DPR | |
|---|---|---|---|---|---|---|---|---|
| VB | MCMC | |||||||
| MFP | 2.26 (0.49) | 3.04 (0.24) | 5.01 (0.75) | 0.27 (0.05) | 0.40 (0.12) | 0.22 (0.11) | 6.29 (3.07) | |
| Cattle | MY | 2.51 (0.52) | 2.95 (0.31) | 5.95 (1.04) | 0.27 (0.08) | 0.46 (0.07) | 0.21 (0.09) | 4.01 (0.55) |
| SCS | 4.56 (0.78) | 3.15 (0.27) | 6.17 (1.05) | 0.24 (0.04) | 0.27 (0.06) | 0.20 (0.08) | 5.23 (2.38) | |
| Maize | GDD | 2.38 (0.72) | 1.08 (0.11) | 7.86 (1.57) | 0.19 (0.05) | 0.03 (0.01) | 0.08 (0.01) | 4.53 (1.29) |
| LDL | 1.02 (0.17) | 1.78 (0.15) | 78.56 (27.78) | 1.76 (1.15) | 1.71 (0.33) | 1.24 (0.79) | 85.76 (18.22) | |
| GLU | 0.25 (0.14) | 1.86 (0.18) | 47.87 (17.86) | 1.06 (0.52) | 1.63 (0.13) | 0.43 (0.12) | 61.16 (23.46) | |
| HDL | 0.49 (0.16) | 1.83 (0.14) | 80.45 (38.23) | 3.39 (1.26) | 1.74 (0.11) | 1.28 (0.56) | 84.38 (10.61) | |
| FHS | TC | 0.24 (0.13) | 1.92 (0.12) | 51.17 (16.72) | 1.05 (0.48) | 1.62 (0.37) | 0.42 (0.11) | 51.69 (11.77) |
| TG | 0.25 (0.17) | 1.98 (0.15) | 59.41 (17.72) | 0.99 (0.35) | 1.91 (0.46) | 0.45 (0.13) | 50.78 (10.72) | |
| Height | 0.68 (0.16) | 1.75 (0.16) | 71.14 (13.80) | 2.27 (1.12) | 4.13 (1.18) | 1.56 (0.18) | 71.62 (11.89) | |
| Weight | 0.59 (0.13) | 1.61 (0.15) | 72.66 (12.15) | 2.28 (1.11) | 1.95 (0.34) | 1.61 (0.10) | 79.67 (15.04) | |
| BMI | 0.47 (0.10) | 1.71 (0.13) | 76.08 (15.28) | 2.31 (1.13) | 2.35 (0.27) | 1.57 (0.17) | 75.15 (14.91) | |
The computational time is in hours. Values in parentheses are standard deviations. Mean and standard deviation are calculated based on 20 replicates. For MCMC-based methods (rjMCMC, BVSR, BayesR, and DPR.MCMC), the computational time is based on 50,000 iterations of Metropolis Hastings steps for BVSR, reversible jump steps for rjMCMC, and Gibbs steps for BayesR and DPR.MCMC