Table 2.
Randomization test of technical clustering software
| Threshold | Original cluster | Weight | Cluster after randomizing Fx4_CM10_high_C066628 | Weight |
|---|---|---|---|---|
| 0.15 |
Fx4_CM10_high_C066628_ |
0.581 |
Not clustered |
NA |
| |
Fx4_CM10_high_C033359_ |
0.419 |
|
|
| 0.20 |
Fx4_CM10_high_C0133296 |
0.288 |
Not clustered |
NA |
| |
Fx4_CM10_high_C066628_ |
0.260 |
|
|
| |
Fx4_CM10_high_C033359_ |
0.176 |
|
|
| |
Fx3_CM10_high_C033347_ |
0.129 |
|
|
| |
Fx4_CM10_high_C022248_ |
0.061 |
|
|
| |
Fx3_CM10_high_C066617_ |
0.053 |
|
|
| |
Fx3_CM10_high_C022244_ |
0.033 |
|
|
| 0.25 |
Fx4_CM10_high_C0133296 |
0.250 |
Not clustered |
NA |
| |
Fx4_CM10_high_C066628_ |
0.230 |
|
|
| |
Fx4_CM10_high_C033359_ |
0.154 |
|
|
| |
Fx3_CM10_high_C033347_ |
0.121 |
|
|
| |
Fx3_CM10_high_C0133108 |
0.089 |
|
|
| |
Fx4_CM10_high_C022248_ |
0.055 |
|
|
| |
Fx3_CM10_high_C066617_ |
0.051 |
|
|
| |
Fx3_CM10_high_C022244_ |
0.027 |
|
|
| |
Fx4_H50_high_C022244_ |
0.024 |
|
|
| 0.3 |
Fx4_CM10_high_C0133296 |
0.238 |
Fx3_CM10_high_C033347_ |
0.237 |
| |
Fx4_CM10_high_C066628_ |
0.224 |
Fx4_CM10_high_C033359_ |
0.237 |
| |
Fx4_CM10_high_C033359_ |
0.151 |
Fx3_CM10_high_C0133108 |
0.152 |
| |
Fx3_CM10_high_C033347_ |
0.119 |
Fx4_CM10_high_C022248_ |
0.106 |
| |
Fx3_CM10_high_C0133108 |
0.087 |
Fx3_CM10_high_C066617_ |
0.093 |
| |
Fx4_CM10_high_C022248_ |
0.054 |
Fx5_H50_high_C022236_ |
0.065 |
| |
Fx3_CM10_high_C066617_ |
0.050 |
Fx4_CM10_high_C066628_ |
0.056 |
| |
Fx5_H50_high_C022236_ |
0.030 |
Fx3_CM10_high_C022244_ |
0.028 |
| |
Fx3_CM10_high_C022244_ |
0.026 |
Fx4_H50_high_C022244_ |
0.026 |
| |
Fx4_H50_high_C022244_ |
0.020 |
|
|
| 0.4 |
Fx4_CM10_high_C0133296 |
0.238 |
Fx4_CM10_high_C0133296 |
0.240 |
| |
Fx4_CM10_high_C066628_ |
0.224 |
Fx4_CM10_high_C033359_ |
0.165 |
| |
Fx4_CM10_high_C033359_ |
0.151 |
Fx3_CM10_high_C033347_ |
0.154 |
| |
Fx3_CM10_high_C033347_ |
0.119 |
Fx3_CM10_high_C0197893 |
0.117 |
| |
Fx3_CM10_high_C0133108 |
0.087 |
Fx3_CM10_high_C0133108 |
0.112 |
| |
Fx4_CM10_high_C022248_ |
0.054 |
Fx3_CM10_high_C066617_ |
0.065 |
| |
Fx3_CM10_high_C066617_ |
0.050 |
Fx4_CM10_high_C022248_ |
0.064 |
| |
Fx5_H50_high_C022236_ |
0.030 |
Fx5_H50_high_C022236_ |
0.036 |
| |
Fx3_CM10_high_C022244_ |
0.026 |
Fx3_CM10_high_C022244_ |
0.024 |
| |
Fx4_H50_high_C022244_ |
0.020 |
Fx4_H50_high_C022244_ |
0.012 |
| |
|
|
Fx4_CM10_high_C066628_ |
0.012 |
| 0.5 |
Fx4_CM10_high_C0133296 |
0.161 |
Fx4_CM10_high_C0133296 |
0.194 |
| |
Fx4_CM10_high_C066628_ |
0.140 |
Fx4_CM10_high_C0196854 |
0.155 |
| |
Fx4_CM10_high_C0196854 |
0.129 |
Fx4_CM10_high_C033359_ |
0.101 |
| |
Fx4_CM10_high_C033359_ |
0.088 |
Fx3_CM10_high_C033347_ |
0.074 |
| |
Fx3_CM10_high_C033347_ |
0.064 |
Fx3_CM10_high_C0197893 |
0.065 |
| |
Fx6_CM10_high_C033435_ |
0.060 |
Fx6_CM10_high_C033435_ |
0.065 |
| |
Fx3_CM10_high_C0197893 |
0.055 |
Fx3_CM10_high_C0133108 |
0.061 |
| |
Fx3_CM10_high_C0133108 |
0.052 |
Fx6_CM10_high_C066702_ |
0.042 |
| |
Fx6_CM10_high_C066702_ |
0.036 |
Fx6_CM10_high_C0133574 |
0.039 |
| |
Fx6_CM10_high_C0133574 |
0.033 |
Fx3_H50_high_C033429_ |
0.038 |
| |
Fx3_H50_high_C033429_ |
0.032 |
Fx4_CM10_high_C022248_ |
0.034 |
| |
Fx4_CM10_high_C022248_ |
0.030 |
Fx3_CM10_high_C066617_ |
0.033 |
| |
Fx3_CM10_high_C066617_ |
0.028 |
Fx3_H50_high_C066672_ |
0.023 |
| |
Fx3_H50_high_C066672_ |
0.019 |
Fx5_H50_high_C022236_ |
0.020 |
| |
Fx5_H50_high_C022236_ |
0.018 |
Fx3_H50_high_C0133496 |
0.019 |
| |
Fx3_CM10_high_C022244_ |
0.016 |
Fx3_CM10_high_C022244_ |
0.018 |
| |
Fx3_H50_high_C0133496 |
0.016 |
Fx4_H50_high_C022244_ |
0.017 |
| |
Fx4_H50_high_C022244_ |
0.013 |
Fx4_CM10_high_C066628_ |
0.002 |
| |
Fx5_CM10_high_C0133649 |
0.010 |
|
|
| Fx5_CM10_high_C066776_ | 0.001 |
The Technical Clustering software was applied to the dataset of spectra obtained from 85 samples before (left-hand column) and after (right hand column) randomizing the sample identities of the intensities of one peak (Fx4_CM10_high_C066628, in bold type). The clusters that contain the correct sample values are shown in the “Original Cluster” column at the indicated threshold CV, and after randomizing the sample intensities of just one peak on the right. The contributions (weights) of the components of the clusters are also shown.