CPPTRAJ in AMBER24 incorrectly calculates the CA-CA distance between amino acid residues in a protein.
I used the AMBER24 molecular dynamics simulation of a large protein containing 1488 residues. The initial conformation of the protein is very extended. After several standard geometry optimization steps in a periodic rectangular water box also containing Na+ Cl- ions at physiological concentration, I used a short conventional molecular dynamics simulation of 5 ns duration to equilibrate the whole system:
equilibrate 5ns
&cntrl
imin=0,irest=1,ntx=5,
nstlim=2500000,dt=0.002,
ntc=2,ntf=2,ig=-1,
cut=8.0, ntb=2, ntp=1, taup=2.0,
ntpr=1000, ntwx=1000,
ntt=3, gamma_ln=2.0,
temp0=310.0,
/
In the next step, I used the CPPTRAJ script cpptr_dist.in to calculate the CA-CA distance value between residues ASP923 and ALA1479 in the last 2500 snapshot:
# Distance analysis with cpptraj.
# Load topology trajectory
parm ../mc.prmtop
trajin ../6_/eq.mdcrd 2500 2500
strip :WAT,Cl-,Na+
autoimage origin
distance ASP923_ALA1479 :923.CA :1479.CA out dist_CA-CA_ASP923_ALA1479_eq_2500.agr
run
The result is as follows:
with g0
xaxis label "Frame"
yaxis label ""
legend 0.2, 0.995
legend char size 0.60
s0 legend "ASP923_ALA1479"
target G0.S0
type xy
1.000 139.327
I then used the CPPTRAJ script cpptr_pdb.in to generate a PDB file (If needed I'll send the PDB file by request) from the RST file containing the final 3D structure after the 5 ns MD simulation:
parm ../mc.prmtop
trajin eq.rst
autoimage origin
trajout eq_5ns.pdb pdb
run
quit
The next two lines contain the XYZ coordinates of the two CA target atoms copied from the PDB file:
ATOM 14005 CA ASP 923 -52.292 65.593 25.838
ATOM 22425 CA ALA 1479 116.039 -138.392 -131.288
The target CA-CA distance calculated from PDB coordinates is 307.626 A. The corresponding CA-CA distance value generated by CPPTRAJ is 139.327 A, which is completely wrong!
There seems to be a bug in the CPPTRAJ software.
I need to use CPPTRAJ to analyze CA-CA distance variations during long
MD production simulations.
I would appreciate help from AMBER experts in fixing this issue.
Thank you,
Michael
*****************************
Michael Shokhen, PhD
Associate Professor
Department of Chemistry
Bar Ilan University,
Ramat Gan, 52900
Israel
email: michael.shokhen.biu.ac.il<https://webmail.biu.ac.il/owa/redir.aspx?C=a160ef9b9a6b4d06992402715d3ee465&URL=mailto%3ashokhen%40mail.biu.ac.il>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri May 16 2025 - 09:00:02 PDT
