Table 2.

Comparison of MCLUST, MFA, MCFA and MCtFA models for implied clustering versus the true membership of Bhattacharjee's 203 lung cancer tissues

ModelFactorsBICAWEARIError rate

MCLUSTVVI135 0230.30210.3350
MFA1140 710145 3240.31000.4286
2139 479146 1250.32190.3941
3139 313147 9550.31560.4089
4139 016149 6110.30130.4039
5139 068151 5730.33680.3645
6138 870153 2440.24450.4236
7139 358155 5610.34870.3498
8139 747157 7380.26160.4335
9140 207159 9430.25710.4433
10140 414161 8540.23680.4680
MCFA1148 255149 6740.0721NA
2144 994146 5850.33480.2611
3142 978145 0420.30900.3300
4141 826144 4530.43760.2759
5140 943144 1390.37030.3153
6140 123143 9430.32690.3251
7139 362143 8000.36920.3153
8138 921144 0150.37750.3153
9138 420144 1940.45700.2709
10138 134144 6330.24180.4335
MCtFA1144 424145 607−0.12690.4384
2142 708144 2940.36190.2413
3141 155143 2360.47350.2266
4139 683142 3340.61790.1527
5138 937142 1540.66570.1379
6138 194142 0250.73220.1133
7137 538142 0090.58750.1675
8136 973142 1050.64170.1773
9136 660142 4740.44080.2365
10136 431142 9670.32150.3153
ModelFactorsBICAWEARIError rate

MCLUSTVVI135 0230.30210.3350
MFA1140 710145 3240.31000.4286
2139 479146 1250.32190.3941
3139 313147 9550.31560.4089
4139 016149 6110.30130.4039
5139 068151 5730.33680.3645
6138 870153 2440.24450.4236
7139 358155 5610.34870.3498
8139 747157 7380.26160.4335
9140 207159 9430.25710.4433
10140 414161 8540.23680.4680
MCFA1148 255149 6740.0721NA
2144 994146 5850.33480.2611
3142 978145 0420.30900.3300
4141 826144 4530.43760.2759
5140 943144 1390.37030.3153
6140 123143 9430.32690.3251
7139 362143 8000.36920.3153
8138 921144 0150.37750.3153
9138 420144 1940.45700.2709
10138 134144 6330.24180.4335
MCtFA1144 424145 607−0.12690.4384
2142 708144 2940.36190.2413
3141 155143 2360.47350.2266
4139 683142 3340.61790.1527
5138 937142 1540.66570.1379
6138 194142 0250.73220.1133
7137 538142 0090.58750.1675
8136 973142 1050.64170.1773
9136 660142 4740.44080.2365
10136 431142 9670.32150.3153

The bold numbers are the optimal values of BIC, AWE, ARI and Error rate for each model. NA means Not Available.

Table 2.

Comparison of MCLUST, MFA, MCFA and MCtFA models for implied clustering versus the true membership of Bhattacharjee's 203 lung cancer tissues

ModelFactorsBICAWEARIError rate

MCLUSTVVI135 0230.30210.3350
MFA1140 710145 3240.31000.4286
2139 479146 1250.32190.3941
3139 313147 9550.31560.4089
4139 016149 6110.30130.4039
5139 068151 5730.33680.3645
6138 870153 2440.24450.4236
7139 358155 5610.34870.3498
8139 747157 7380.26160.4335
9140 207159 9430.25710.4433
10140 414161 8540.23680.4680
MCFA1148 255149 6740.0721NA
2144 994146 5850.33480.2611
3142 978145 0420.30900.3300
4141 826144 4530.43760.2759
5140 943144 1390.37030.3153
6140 123143 9430.32690.3251
7139 362143 8000.36920.3153
8138 921144 0150.37750.3153
9138 420144 1940.45700.2709
10138 134144 6330.24180.4335
MCtFA1144 424145 607−0.12690.4384
2142 708144 2940.36190.2413
3141 155143 2360.47350.2266
4139 683142 3340.61790.1527
5138 937142 1540.66570.1379
6138 194142 0250.73220.1133
7137 538142 0090.58750.1675
8136 973142 1050.64170.1773
9136 660142 4740.44080.2365
10136 431142 9670.32150.3153
ModelFactorsBICAWEARIError rate

MCLUSTVVI135 0230.30210.3350
MFA1140 710145 3240.31000.4286
2139 479146 1250.32190.3941
3139 313147 9550.31560.4089
4139 016149 6110.30130.4039
5139 068151 5730.33680.3645
6138 870153 2440.24450.4236
7139 358155 5610.34870.3498
8139 747157 7380.26160.4335
9140 207159 9430.25710.4433
10140 414161 8540.23680.4680
MCFA1148 255149 6740.0721NA
2144 994146 5850.33480.2611
3142 978145 0420.30900.3300
4141 826144 4530.43760.2759
5140 943144 1390.37030.3153
6140 123143 9430.32690.3251
7139 362143 8000.36920.3153
8138 921144 0150.37750.3153
9138 420144 1940.45700.2709
10138 134144 6330.24180.4335
MCtFA1144 424145 607−0.12690.4384
2142 708144 2940.36190.2413
3141 155143 2360.47350.2266
4139 683142 3340.61790.1527
5138 937142 1540.66570.1379
6138 194142 0250.73220.1133
7137 538142 0090.58750.1675
8136 973142 1050.64170.1773
9136 660142 4740.44080.2365
10136 431142 9670.32150.3153

The bold numbers are the optimal values of BIC, AWE, ARI and Error rate for each model. NA means Not Available.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close