@hackage / PDBtools

These modules provide an algebraic data type for atoms and proteins. The data is expected to come from Protein Data Bank .pdb files. The parser will interpret ATOM and HETATM record types.

When implementing the parser over a large number of files:

1. Compile your code.
2. Use mapM_ in your main loop to ensure computations are done sequentially, otherwise you'll likely see a stack overflow. There are certainly other solutions to memory management, depending on the sorts of computations that you're hoping to do. If mapM_ doesn't suit your goals, you may want to explore seq and deepSeq.

At the moment, all vector operations are conducted by Vectors.hs. Be warned that this is an unsafe module (it makes no effort to check vector lengths, etc. In our case everything is three dimensional, so this isn't a problem. Vectors.hs is a module that could benefit from a great deal of expansion. Matrix operations are hopefully coming soon.