Couple newbie questions: 1. I don't know if any users have run RaptorX (U-Chicago) for ligand-residue pocket prediction, but the results simply tell you the specific AA residues are which are close to the pocket For example the locations could simply be e.g. the residues K252, Q256, and K260 in the input PDB file. So my question in terms of AA annotation in Chimera is, how could I center Autodock Vina's pocket box around residues something like K252, Q256, and K260? I looked through the select options and didn't see where numbered residues could be selected. 2. For 3D modeling when x-ray crystallography results are not available in PDB, I commonly will submit a protein's FASTA sequence to SWISS-MODEL, and then download the PDB and go through dock prep for the receptor and minimization of the ligand(and adding e.e. hydrogens, etc.) Isn't the SWISS-MODEL approach an alternative for using the MODELLER add-on for homology modeling. The MODELLER run I just went through was really using a lot of core resources, and the BLAST template results were complexes between receptors and ligands, and sometimes DNA. So isn't there an advantage to inputting a FASTA from Uniprot into SWISS-MODEL to obtain the predicted 3D structure, when compared with the BLAST results required during MODELLER runs?