Thank for providing a sample PDB file which helps clarify the issue.
ATOM 5 O1P CYT A 1 17.752 35.775 26.807 1.00 1.70 N01A
ATOM 6 O2P CYT A 1 17.012 35.542 24.381 1.00 1.70 N01A
ATOM 7 O5' CYT A 1 18.906 34.140 25.293 1.00 1.77 N01A
ATOM 8 O5T CYT A 1 19.074 36.580 24.990 1.00 1.77 N01A
Yes, it is the O5T [8, atom serial number] which is taken as O5' by the '
analyze' program. Since the CYT #1 on chain A already has an O5' atom [7], that explains the message:
more than one O5' atoms : residue name 'CYT', chain A, number [ 1 ]
*****the first atom is used*****
In your case, the assumption of taking the 1st O5' atom [7] in calculations works fine.
Which software did you use? This is the first time I noticed such an O5T naming. Presumably, O5T [8] should be the O3' atom in the preceding (5') nt. Here O5' [7] could be named O5T, which 3DNA handles without any problem.
HTH,
Xiang-Jun