Different units in analysis.LinearDensity

[lilyminium]

Expected behavior

MDAnalysis keeps consistent units of Angstrom for length, or documents when it doesn't.

Actual behavior

analysis.LinearDensity converts volumes from A^3 to cm^3

Code to reproduce the behavior

mdanalysis/package/MDAnalysis/analysis/lineardensity.py

Line 177 in 009ec22

k = 6.022e-1 # divide by avodagro and convert from A3 to cm3

This isn't in the docs.

Currently version of MDAnalysis

• Which version are you using? (run python -c "import MDAnalysis as mda; print(mda.__version__)")
• Which version of Python (python -V)?
• Which operating system?

[lilyminium]

This also makes the header written in the file inaccurate. (charges are indicated to have e/A^3 density)

        header = ("1 coord [Ang] 2-7 mass density (x,sx,y,sz,z,sz) [g/cm^3]"
                  "8-13 charge density (x,sx,y,sz,z,sz) [e/A^3]\n Average "
                  "density: {} g/cm3".format(density))

Edit: the charges are not actually in e, they're in e*(Avogadro's number). Possibly the line below is unintentional.

mdanalysis/package/MDAnalysis/analysis/lineardensity.py

Line 191 in 009ec22

self.results[dim]['char'] /= self.results[dim]['slice volume'] * k

[orbeckst]

@mattihappy can you clarify what was intended?

[orbeckst]

I would be in favor of using standard units, Å, and e/ų.

For density, g/cm³ is more sensible than u/ų.

Note that the save functions were removed #1745 (specifically PR #2487) so this boils down to

1. documenting what is actually done
2. deciding to change the output data without deprecation — do we want to and if so, what should be changed?