Atom names purely in RDKit

For some applications, such as PyMOL scripts or Rosetta, atom names are really important, say CA is the standard name for the α-carbon. Example uses of atom names in Rosetta/pyrosetta include setting constraints, using a params file for a custom ligand and so forth. However, RDKit is a bit of a nuisance with atom names as it is not a central feature, but a feature added for PDB files that is not too well documented.

Working around segmentation faults of pyrosetta: threads & processes

Rosetta often does not die gracefully. Pyrosetta is the same. If the starting template is not great segmentation faults will result in the kernel issuing signal 11 to kill the process. The way around it is to spin it up as its own process via the multiprocessing module and not the threading module, because child threads use the same process.

Guess bond order in Rdkit by number of bound atoms

Some compChem/Biochem programs do not care about bond order and strip them, which is rather frustrating. Ligands in random PDB files without any name, smiles are a classic example.
There is no single magic mol.CorrectBondOrder() command in Rdkit, but luckily there are some tricks that can be done. Here I will discuss finding out using the number of bound atoms.

What-if: Biosynthesis of deazaguanine

7-deazaguanine is an analogue of guanine that has a carbon instead of a nitrogen. This molecule that is not made in nature, but many substituted versions, namely queine, archaeosine nucleobase and their precursors. These are made via a different route. However, it would be very feasible to make via straightforward enzymology.

Convert Python docstrings to GitHub markdown readmes

This is suited for very small projects were read-the-docs is way overkill, however for most project that will be more suited — see how-to post on setting Sphinx for RTD for more

The Greek philosopher Epictetus said that a day reverse engineering a piece of code saves you half an hour reading the documentation. A maxim still valid to this day. Nevertheless, documenting code is important. With PyCharm and the push towards typehinting in Python writing docstrings is fairly simple. However, getting docstrings into the readme.md of GitHub is not straightforward the first time round. Hence, I wrote this simple guide to doing so.
Do note that for medium/big projects, using ReadTheDocs is recommended over this hack: ee the guide to setting up Sphinx for ReadTheDocs

Go away glycerol!!

Due to the nature of crystallisation additives are often found in PDB structures. These are generally unwelcome, especially if you want to extract ligands. In fact, I have heard only once someone talk excitedly about their crystallisation reagent in their structure, but only because they were trying to flog it off as an allosteric binding site. Generally, they are just annoying. Luckily you don't need reinvent the wheel as a list or two already exist!

RDKit for Rosetta: PLP ligand space as an example

Docking requires a molecule to dock. Preparing a ligand is often tricky, especially if the ligand is complicated, such as PLP. PLP is an interesting cofactor as it catalyses the reaction while the protein chooses the ligand. It binds tightly to the active site via its phosphate and its pyridine ring, while the metabolite to be transformed forms a Schiff base with it. Therefore, one would think that it makes easy to explore chemistry space with it. However, several technical hurdles are encountered, making it quite didactic.