Table 2.
Pairwise comparisons of different mouse HD classifications: VI distances
| SCI-PHYEntropy mH178 | SCI-PHYEntropy H559 | CD-HIT | OSLPID | OSLTRE mH178 | OSLTRE H559 | OSLEval mH178 | OSLEval H559 | |
|---|---|---|---|---|---|---|---|---|
| SCI-PHYEntropy mH178 | 0.00 | 0.53 | 0.50 | 0.46 | 0.56 | 0.41 | 0.55 | 0.56 |
| SCI-PHYEntropy H559 | 0.00 | 0.13 | 0.14 | 0.11 | 0.14 | 0.18 | 0.19 | |
| CD-HIT | 0.00 | 0.09 | 0.10 | 0.11 | 0.15 | 0.16 | ||
| OSLPID | 0.00 | 0.10 | 0.05 | 0.10 | 0.10 | |||
| OSLTRE mH178 | 0.00 | 0.15 | 0.09 | 0.10 | ||||
| OSLTRE H559 | 0.00 | 0.14 | 0.15 | |||||
| OSLEval mH178 | 0.00 | 0.01 | ||||||
| OSLEval H559 | 0.00 |
Summary of the pairwise similarity levels, in terms of Variation-of-Information (VI) distances, between the seven different fine-grained sequence-based classifications.
These classifications were derived from the mH178 or H559 data set, as indicated (see ‘Materials and Methods’ section for details), using three different methods: SCI-PHY in the Entropy mode, CD-HIT and the three Objective Single Linkage (OSL) procedures. OSLPID, and CD-HIT each produced exactly the same classification when applied to either of the 2 HD data sets.
Relatively high VI values corresponding to poor similarity levels between the listed classifications are shown as underlined italic.