A Survey on Identification of Protein Complexes in Protein–protein Interaction Data: Methods and Evaluation - Computational Intelligence Techniques for Comparative Genomics - page 62

each attachment in the set, we select a best one with the highest value of “ close-

ness ” as its potential protein complex core. The “ closeness ” of an attachment V to a

core C i is

X

CL ð V ; C i Þ¼ ECC ð v ; u Þ ð 13 Þ

u 2 C l

of an attachment to all the possible protein complex cores is
equal to 0, this attachment will be ignored. For each time progress in the molecular
cycle, every protein will be judged whether it is active or not. If a protein is active in
a certain time course, it will be considered as an attachment and added to its
corresponding protein complex core at this time course. Finally, the protein com-
plex cores and their attachments with the expressed time course will be outputted.

If the “ closeness ”

Filtering False-positive Complexes

In the last stage, it will reexamine the possible dynamic protein complexes to lter

some false positives. According to the formation and function of a protein complex,
it should be active in two or more continual time courses. It uses some of the rules
to

lter false-positive complexes.

1. A protein complex should include at least two proteins.
2. The attachment proteins should be active in the same time course or in different
but adjacent time courses.
3. If the attachments of a possible protein complex do not satisfy the second rule
and the protein complex core involves at least two proteins, the core will be kept
as a nal protein complex.

Finally, one can identify the protein complex with these stages, namely detecting
protein complex cores, generating potential dynamic complexes, and ltering false-

positive complexes.

3 Computational Analysis of Protein Complexes

To evaluate the clustering techniques against explicit datasets that previously
contain information about recorded protein complexes [ 21 ]. For comprehensive
evaluation, we employed several valuation measures with co-annotation, co-

localization, precision, recall, F-measure, sensitivity, positive predictive value, and
separation values.

“

”