All the cases were analysed anonymously, that is, all the specimens were assigned a new number without any personal information.
With the exception of three individuals, all the remaining 142 (98%) specimens were assigned in to the same clusters as in the previous analysis, indicating that subdivision was not locus-dependent.
This accuracy increased when samples were distinguished geographically: 96% of all fresh specimens and 99.5% of unidentified museum specimens were assigned to the same morphotype as with our manual method.
Although originally twelve specimens were assigned to the taxon, at least one of them may belong to a different genus.
* At this threshold, only 131 specimens were assigned to classes.
Hence, the specimens were assigned to the "adult" category.
Isolates recovered from patient specimens were assigned arbitrary numbers.
Specimens were assigned to species and molecular form using the PCR-RFLP method developed by Fanello et al. (2002).
Female specimens were assigned to species by association with males and according to body dimensions and colouration.
The specimens were assigned to G. tigrina on the basis of their external appearance [ 5, 14, 15].
Specimens were assigned scores of 0 3 based on an ordinal scale representing increasing loss of subcellular localization and polarity.
