Standard file format from pdb database. Server uses all atoms. Define chain index in submit form if your file contains more than one chain.
GapRepairer
[...]
ATOM 350 CE1 PHE A 41 -9.807 79.442-138.392 1.00 14.73 C
ATOM 351 CE2 PHE A 41 -8.711 78.736-140.419 1.00 13.64 C
ATOM 352 CZ PHE A 41 -8.731 78.828-139.029 1.00 11.73 C
ATOM 353 N THR A 42 -13.650 83.537-142.079 1.00 11.61 N
ATOM 354 CA THR A 42 -14.651 84.121-142.964 1.00 11.91 C
ATOM 355 C THR A 42 -13.939 85.104-143.886 1.00 11.88 C
ATOM 356 O THR A 42 -12.832 85.552-143.582 1.00 10.10 O
ATOM 357 CB THR A 42 -15.714 84.903-142.180 1.00 11.18 C
ATOM 358 OG1 THR A 42 -15.069 85.916-141.403 1.00 10.45 O
ATOM 359 CG2 THR A 42 -16.509 83.969-141.264 1.00 8.81 C
ATOM 360 N PHE A 43 -14.580 85.443-145.000 1.00 10.49 N
ATOM 361 CA PHE A 43 -14.001 86.377-145.958 1.00 10.78 C
ATOM 362 C PHE A 43 -14.887 87.571-146.281 1.00 12.52 C
ATOM 363 O PHE A 43 -14.427 88.531-146.896 1.00 11.30 O
ATOM 364 CB PHE A 43 -13.656 85.652-147.257 1.00 8.24 C
ATOM 365 CG PHE A 43 -12.577 84.630-147.102 1.00 8.81 C
ATOM 366 CD1 PHE A 43 -12.881 83.273-147.079 1.00 7.91 C
ATOM 367 CD2 PHE A 43 -11.249 85.027-146.974 1.00 9.09 C
ATOM 368 CE1 PHE A 43 -11.873 82.323-146.933 1.00 8.49 C
[...]
To repair an incorrect structure (or e.g. modify a loop from a complete structure to resemble fold from its homologue) you should first remove the offending residues from the PDB file. To ensure that the new structure isn't based on the starting one (if it is published in PDB database) you should then either exclude it (Exclude structures under the advanced options tab) or specify the desired templates (if known). If there are no homologues to be found the removed region will be reconstructed de novo using modeller's refinement routine.
When removing residues it is important to delete all of it's atoms - leaving some behind may cause an unpredictable behaviour.
GapRepairer doesn't move the atoms already present in the PDB file. If you believe that residues surrounding the gap have been crystalized incorrectly you should remove them from the input file prior to uploading it. For structures taken from the PDB database you should also exclude them ( specify their PDB ID in Exclude structures under the advanced options tab) to ensure that removed amino acids will not be modelled based on their previous coordinates
When removing residues it is important to delete all of it's atoms - leaving some behind may cause an unpredictable behaviour.