Hi Patrick,

Good eye! These differences are explained in the "notes" section after the description of the experiments on:
http://gofightagainstmalaria.scripps.edu/inde...r-potential-malaria-drugs

For example:
"Since the crystal structures of Pf OAR were missing the coordinates of a flexible loop that is adjacent to the active site, the program called "loopy" (from Barry Honig's lab) was used to create different models of possible conformations of this flexible loop. Loopy can be obtained at http://trantor.bioc.columbia.edu or at http://wiki.c2b2.columbia.edu/honiglab_public/index.php/Software . Different "force fields" were used within loopy to obtain different models of this flexible loop, such as AMBER and CHARMM."


"Note: the crystal structures for FabZ had a few missing loops (that is, those loops were likely too flexible or polymorphic to difract well; thus, the crystal structures contained no coordinates for those loops). The "loopy" tool from the Honig lab was used to create models for these missing loops. If the AMBER force-field was used to make the model within loopy, then "loop_amber" is listed above. Conversely, if the CHARMM force-field was used to make the model within loopy, then "loop_charmm" is listed above)."


For the experiments versus FabZ, I also included one original structure in which the loops were not added. This will allow me to compare the accuracy (i.e., the predictive utility) of these different types of models for that target.

Thank you for your interest and your continued support,
Alex L. Perryman, Ph.D.