Benchmarking Bio.PDB

These scripts allow you to download and benchmark the PDB and mmCIF parser in Biopython (Bio.PDB) on the entire PDB database.

The benchmark scripts will produce a *.xml file for each data file with timings and memory usage, as well as brief details of the chemical composition of the structure (number of chains, atoms, etc).

The parsers are validated by cross-comparison and independent comparison to similar XML output files derived from PDBML files (see download_pdb.sh).

Instructions

pip install psutil  # to track memory usage
python -m pip install git+https://github.com/biopython/biopython.git  # get latest revision
./download_pdb.sh  # takes a few hours

python benchmark_parsers.py pdb pdb/ &> benchmark.pdb.out
python benchmark_parsers.py mmcif mmCIF/ &> benchmark.cif.out
python process_xmldb.py XML &> xmlprocess.out

Questions?

Feel free to send me an email or open an issue here and tag me.