This assessment contained information on health activities, but fell short of the comprehensiveness of the other more spontaneous datasets examined, in large part due to the nature of the 3W tool, which is not generally intended to be used as a health facility assessment tool or census.
In particular, the C-terminal tail was shown to evolve under more relaxed selective constraints in the four datasets examined and many sites pinpointed to be under adaptive evolution were located in this region.
The rank order of the algorithms is limited to the datasets examined in this study, however, and comparisons of the algorithms across more diseases/conditions and persons with different demographic characteristics may provide stronger evidence of the rank order "stability".
Of all datasets examined, the laptop dataset showed the best classification accuracy (98.86%).
Further, with more sequences and the factorial increase in the number of possible trees it is desirable to make a corresponding increase in the number of resampled datasets examined.
We show that Normalized-Cut also produces some incorrect results for some of the datasets examined.
These results were consistently observed in the four datasets examined.
In the future, we will continue to examine our techniques in more datasets.
Publication bias was examined in funnel plots and with a regression asymmetry test: the Egger test is best for cancer sites with 10 or more datasets.
More datasets are planned for release, based on feedback.
