Analysis of helminth genetic data: comparative examples with Haemonchus contortus isozymes using exact tests or resampling procedures

Abstract

Population genetic data on helminths are scarce and it is important to fine tune analyses of these data, the interpretation of which is based on between-(F _st) and within-(F _is) population variability. Several computer programs are available and confidence intervals of each index are based on different procedures. These programs (Genetix, Diploid, Genepop) were compared with a classical program (Biosys-1) using a large set of Haemonchus contortus isozyme data. The results were identical for all softwares for large differences when natural populations were studied (F _st>0.25 and F _is>0.15 or <−0.15). When small F _st were investigated in laboratory-reared populations, the significance of differences was often related to the software used. As a rule of thumb, it is suggested that at least two out of three software packages (based on different calculation procedures) should give similar results before differences are validated.