Table 4. The structure of the guide trees generated by different algorithms in terms of the coverage of reference alignments.
| Dataset | single linkage |
UPGMA |
Clustal Omega |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Sackin idx | Ref. frac. | MC frac. | Sackin idx | Ref. frac. | MC frac. | Sackin idx | Ref. frac. | MC frac. | |
| small | 147.09 | 0.624 | 0.835 | 21.30 | 0.691 | 0.879 | 18.71 | 0.726 | 0.882 |
| medium | 457.50 | 0.627 | 0.837 | 28.63 | 0.714 | 0.883 | 24.76 | 0.726 | 0.893 |
| large | 941.30 | 0.681 | 0.842 | 31.13 | 0.675 | 0.885 | 25.90 | 0.772 | 0.895 |
| extra-large | 3,440.74 | 0.750 | 0.857 | 36.09 | 0.697 | 0.901 | 29.39 | 0.839 | 0.910 |
| All | — | 0.662 | 0.841 | — | 0.694 | 0.886 | — | 0.758 | 0.893 |
Column “Sackin idx” presents the normalized Sackin indexes. Column “Ref. frac.” provides the fractions of guide trees covering all reference sequences. Column “MC frac.” Shows the fractions of guide trees covering randomly selected subsets of sequences, averaged over 1000 trials.