Multiwfn forum

Multiwfn official website: http://sobereva.com/multiwfn. Multiwfn forum in Chinese: http://bbs.keinsci.com/wfn. E-mail of admin: sobereva[at]sina.com

You are not logged in.

#1 2021-07-07 23:00:33

sobereva
Tian Lu (Multiwfn developer)
From: Beijing
Registered: 2017-09-11
Posts: 1,624
Website

NCI analysis for protein-ligand

Today a Multiwfn user asked me how to perform NCI analysis for interaction between protein-ligand using geometry generated by molecular docking. My reply may be also useful for other Multiwfn users, so I paste it below:

You can perform NCI analysis using Multiwfn based on the .pdb file containing docked geometry, but I suggest performing an energy minimization using molecular force field first, the resulting map will be more realistic. Note that since protein+ligand is usually a large system and quantum chemistry calculation is usually difficult, commonly you need to use promolecular approximation form of NCI (promolecular NCI).

In Section 5 of my blog article http://sobereva.com/591 (written in Chinese), there is aNCI (averaged NCI) map showing interaction between ligand and truncated protein. You can also plot a similar map using promolecular NCI. (Using aNCI as described in Section 3.23.3 of Multiwfn manual is in principle better, since it can take time average into account, but you need to perform molecular dynamics simulation first, so the computational cost is much higher).

It is worth to note that you should not directly use the .pdb file of "ligand + entire protein" as input file, because there are too many atoms in protein, the computational cost of promolecular NCI prohibitively high. You should use VMD program to extract a cluster containing the ligand with surrounding ligands, like the model shown in Section 5 of http://sobereva.com/591. Specificially, in the "File" - "Save Coordinate" panel of VMD, you can input "resname MOL or protein same resid as within 3.5 of resname MOL" as selected atoms, the molecule with residue name "MOL" and all protein atoms with in 3.5 Angstrom of the molecule will be saved to a new pdb file, which can then be used as input file for NCI analysis.

I highly suggest using IGM instead of promolecular NCI to study the ligand-protein interaction. IGM map is somewhat similar to promolecular NCI map, its major advantage is that you can clearly specify which two fragments to study the interaction between, and the isosurfaces corresponding to interactions of other areas will not obstruct your sight. See Section 3.23.5 of manual about basic knowledge of IGM method and Section 4.20.10 for examples. Another advantage of IGM over promolecular NCI is that the isosurface of IGM is much smoother. The input file for IGM analysis is exactly identical to promolecular NCI analysis.

A very important point of performing IGM or NCI analysis for ligand-protein interaction in Multiwfn is properly defining the grid setting. I suggest using GUI interface (i.e. option "10 Set box of grid data visually using a GUI window") in Multiwfn to easily set up the box for calculating, the box should only cover the interaction region of interest to avoid wasting computation, a grid spacing of 0.2 or 0.25 Bohr is usually sufficient for NCI, while larger spacing is also acceptable for IGM (the smaller the grid spacing, the higher the cost). An example of using this mode of setting up grid is shown in part 3 of my NCI video tutorial: https://youtu.be/e4FpVc9ao48.

Offline

Board footer

Powered by FluxBB