Evaluation of algorithmic methods
In order to get comprehensive picture of strength and weaknesses of algorithmic methods we evaluate their performance using Balanced_Domain_Benchmark_2 dataset and a broad array of criteria such as extend of domain fragmentation, extend of cuts through secondary structure, etc. The results are presented in graphical and tabular form below.
Criteria for evaluation:
Summary of performance of each automatic method.
Composite evaluation of methods. Three characteristics are weighted as follows: fraction of correctly assigned chains: 60%, fragmentation of discontinuous domains: 20%, precision of domain overlap:20%. The overall performance is represented in percent.
This work is sponsored by the National Institutes of Heath (NIH) Grant Number GM63208 (NIH/NIGMS)