PDBeFold Comparisons.

Material from this page cannot be reproduced without the permission of the authors.

The following materials are intended to give you an impression of how PDBeFOLD compares with some other services that also deliver protein structure matching in 3D. The compared characteristics include length of Ca-alignment, achieved RMSD, match index, P-values and Z-scores.

We do not compare the computational performance of different algorithms. PDBeFOLD is a multi-processor parallel system with variable number of employed CPUs, depending on the estimated complexity of a job. It runs on a LSF farm. Other services might be running on single processes or different machines, some of them use pre-matched libraries of structural neighbours or operate on a reduced PDB set. This makes a straightforward comparison of computational efficiency impossible. However, in all instances PDBeFOLD delivers its results in a much shorter time period as compared to any other service, by a factor ranging from 3-5 to 100 and even more. We found that that factor is larger for larger structures.

The comparisons are available for the following PDB entries:

Our findings are summarized in General conclusions