I am currently trying out and comparing different hydrogen bond calculation approaches. The coulomb energy equation of the DSSP algorithm calculates the energy of hydrogen bonds between NH and CO in the backbone as:
Ehb = q1*q2*(1/rNO + 1/rCH - 1/rHO - 1/rCN)*332
where q1, q2 = the charges of N and O, and C is the neighbor atom of the acceptor.
In your opinion, is this model suitable for sidechain-backbone and sidechain-sidechain hydrogen bonds, as well as backbone-backbone?
Also, a more technical question: in the original DSSP approach, charge values for q1 and q2 are 0.42e and 0.20e for N and O, respectively. However, the calculations of charges for my proteins (done in PDB2PQR) give different values for N and O atoms (for example, -0.47 and -0.51). Would the resulting energy values be more accurate if I used these values instead of the ones used in DSSP?
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics