Skip to main content

Table 2 Results of splitting algorithm, CVX and SDPNAL for solving the Block PCA model

From: Inference of gene regulatory subnetworks from time course gene expression data

  

Points30

Points50

Points100

Points200

Points500

funVal

split

146.4654

301.4004

793.5010

2.1104e3

8.0980e3

 

cvx

146.4654

--

--

--

--

 

sdpnal

149.9589

307.0276

794.4018

2.1192e3

8.1217e3

||D - L - E||

split

3.0041e-5

1.0667e-5

7.5166e-5

1.4488e-4

3.7525e-4

 

cvx

1.2590e-6

--

--

--

--

 

sdpnal

4.2797e-5

2.3411e-4

9.1722e-4

0.0033

0.0057

||L||*

split

29.5328

54.8451

118.4007

260.8353

720.8204

 

cvx

29.5323

--

--

--

--

 

sdpnal

29.8736

55.2079

119.9717

261.0082

707.0530

〈W, |L|〉

split

631.1912

1.5986e3

7.0579e3

2.7437e4

1.8056e5

 

cvx

631.1811

--

--

--

--

 

sdpnal

700.7054

1.7995e3

7.0395e3

2.8083e4

1.8329e5

||E||1

split

451.1092

1.2638e3

4.6336e3

1.7925e4

1.1079e5

 

cvx

451.1147

--

--

--

--

 

sdpnal

447.5222

1.2408e3

4.6324e3

1.7854e4

1.1081e5

elapsed time(s)

split

4.88

19.66

80.43

687.65

1.4342e4

 

cvx

6.74

--

--

--

--

 

sdpnal

78.99

265.48

1.3231e3

6.9678e3

5.1222e4

  1. "Points30" indicates the data with 30 points on a plane, and similarly for "Points50", ..., "Points500". "funVal" represents the calculated objective function value of the Block PCA model. "--" indicates out of memory. "2.1104e3" indicates 2.1104 × 103. The implementation of SDPNAL is version 0 [28] and CVX is version 1.21 [29]. The SDPT3 engine is chosen for the CVX solver. The two arguments for the splitting algorithm are set as follows: μ is constantly set 5 for all tests, β is set 20/n for Points30, 200/n for Points50, Points100, Points200, and 500/n for Points500, where n is the number of points.