# gmx nmeig¶

## Synopsis¶

gmx nmeig [-f [<.mtx>]] [-s [<.tpr>]] [-of [<.xvg>]] [-ol [<.xvg>]]
[-os [<.xvg>]] [-qc [<.xvg>]] [-v [<.trr/.cpt/...>]]
[-xvg <enum>] [-[no]m] [-first <int>] [-last <int>]
[-maxspec <int>] [-T <real>] [-P <real>] [-sigma <int>]
[-scale <real>] [-linear_toler <real>] [-[no]constr]
[-width <real>]


## Description¶

gmx nmeig calculates the eigenvectors/values of a (Hessian) matrix, which can be calculated with gmx mdrun. The eigenvectors are written to a trajectory file (-v). The structure is written first with t=0. The eigenvectors are written as frames with the eigenvector number and eigenvalue written as step number and timestamp, respectively. The eigenvectors can be analyzed with gmx anaeig. An ensemble of structures can be generated from the eigenvectors with gmx nmens. When mass weighting is used, the generated eigenvectors will be scaled back to plain Cartesian coordinates before generating the output. In this case, they will no longer be exactly orthogonal in the standard Cartesian norm, but in the mass-weighted norm they would be.

This program can be optionally used to compute quantum corrections to heat capacity and enthalpy by providing an extra file argument -qcorr. See the GROMACS manual, Chapter 1, for details. The result includes subtracting a harmonic degree of freedom at the given temperature. The total correction is printed on the terminal screen. The recommended way of getting the corrections out is:

gmx nmeig -s topol.tpr -f nm.mtx -first 7 -last 10000 -T 300 -qc [-constr]

The -constr option should be used when bond constraints were used during the simulation for all the covalent bonds. If this is not the case, you need to analyze the quant_corr.xvg file yourself.

To make things more flexible, the program can also take virtual sites into account when computing quantum corrections. When selecting -constr and -qc, the -begin and -end options will be set automatically as well.

Based on a harmonic analysis of the normal mode frequencies, thermochemical properties S0 (Standard Entropy), Cv (Heat capacity at constant volume), Zero-point energy and the internal energy are computed, much in the same manner as popular quantum chemistry programs.

## Options¶

Options to specify input files:

-f [<.mtx>] (hessian.mtx)
Hessian matrix
-s [<.tpr>] (topol.tpr)
Portable xdr run input file

Options to specify output files:

-of [<.xvg>] (eigenfreq.xvg)
xvgr/xmgr file
-ol [<.xvg>] (eigenval.xvg)
xvgr/xmgr file
-os [<.xvg>] (spectrum.xvg) (Optional)
xvgr/xmgr file
-qc [<.xvg>] (quant_corr.xvg) (Optional)
xvgr/xmgr file
-v [<.trr/.cpt/…>] (eigenvec.trr)
Full precision trajectory: trr cpt tng

Other options:

-xvg <enum> (xmgrace)
xvg plot formatting: xmgrace, xmgr, none
-[no]m (yes)
Divide elements of Hessian by product of sqrt(mass) of involved atoms prior to diagonalization. This should be used for ‘Normal Modes’ analysis
-first <int> (1)
First eigenvector to write away
-last <int> (50)
Last eigenvector to write away. -1 is use all dimensions.
-maxspec <int> (4000)
Highest frequency (1/cm) to consider in the spectrum
-T <real> (298.15)
Temperature for computing entropy, quantum heat capacity and enthalpy when using normal mode calculations to correct classical simulations
-P <real> (1)
Pressure (bar) when computing entropy
-sigma <int> (1)
Number of symmetric copies used when computing entropy. E.g. for water the number is 2, for NH3 it is 3 and for methane it is 12.
-scale <real> (1)
Factor to scale frequencies before computing thermochemistry values
-linear_toler <real> (1e-05)
Tolerance for determining whether a compound is linear as determined from the ration of the moments inertion Ix/Iy and Ix/Iz.
-[no]constr (no)
If constraints were used in the simulation but not in the normal mode analysis you will need to set this for computing the quantum corrections.
-width <real> (1)
Width (sigma) of the gaussian peaks (1/cm) when generating a spectrum