1. the first dataset we used was collected by Bader and Hogue, please refer the paper by Bader GD, Hogue CWV: "An automated method for finding molecular complexes in large protein interaction networks. BMC Bioinformatics 2003, 4:2. [Evaluation Studies]".

  2. The second Protein interaction data was collected from the MIPS database. The data was publicly available from MIPS protein interaction data.

  3. The third data was collected from The BIOGRID, which is available at BIOGRID protein interaction data.

We evaluate our method against a benchmark dataset of known yeast protein complexes retrieved from the MIPS: Comprehensive Yeast Genome Database . The protein complexes in this dataset have been curated from the biomedical literature.

MIPS function is used in computing the reliability, which is available MIPS functional information . The protein complexes in this dataset have been curated from the biomedical literature.

p-value is computed by using Ontologizer, which is available P-value computation. The protein complexes in this dataset have been curated from the biomedical literature.