In a recent article of mine Identification of new, well-populated amino-acid sidechain rotamers involving hydroxyl-hydrogen atoms and sulfhydryl-hydrogen atoms I look at hydrogen atoms in crystal structures and find that crystallographers can sometimes be really lazy.

Crystallography is the method of working out the molecular structure of proteins, some as large as hundreds of thousands of Daltons (a Dalton is the weight of 1 hydrogen atom). By coaxing your pure sample of powdered protein to form a crystal (dissolving your protein in the right solution), you can shine a big fat xray beam at the crystallized form of your protein, and measure the resultant xray diffraction pattern.

This x-ray diffraction can be mathematically converted into an electron density where the electron density is thick around atoms. However, there an intrinsic limits to this electron density. The thing is, there is hardly ever enough electron density to coalesce around a hydrogen atom so that it comes out clearly. In my article, I show that even with the very highest resolution crystal structures, many crystallographers stick a hydrogen atom on the reported molecular structure even where there is absolutely no density or chemical reason to. Bad bad crystallographers:

However, if you look for hydrogen atoms that have formed hydrogen bonds, you can see the electron density stick to the hydrogen bond. Consequently, you can carefully project the position of these hydrogens, which on the suggestion of a reviewer, I ended up doing manually for each protein structure taking 2 sordid weeks out of my life:

After that, you can show definitively that the hydrogen positions are clustered, giving rise to new rotamers for 4 of the amino acids.