PredictingCharacteristics
Chen, P., Deane, C.M., Reinert G. 2007. Bioinformatics, Vol 23, 17, 2314-2321
All programs (*.pyc) are compiled using Python 2.4.
Run PYC files. Please follow the popup questions and input the corresponing filenames for a successful prediction.
Upcast set of category-category interactions
based on the pairwise protein interactions and protein annotation
pairs of characteristic categories
single characteristic, multiple characteristics (for enhanced method)
Upcast set of triples (triples, triangles, lines) of characteristic categories
based on the triple-wise protein interactions and protein annotation
triples (triples, triangles, lines) of characteristic categories
single characteristic, multiple characteristics (for enhanced method)
Query proteins
A list of proteins to be predicted.
The interactions between these proteins and the others are known. For enhanced methods, at least one characteristics of there proteins is known.
Methods
Pair-based
Triplet-based (network-based)
Example -- Predicting the functions of 5 selected proteins.
Datasets
Target proteins (functions unknown)
Convert classifications from txt file to Python shelve PYC file
Output -- structural classification (shelve), functional classification (shelve)
Construct upcast set of category-category interactions PYC file
Output -- use structure classification, function classification, and both classifications
Construct upcast set of triples (triples, triangles, lines) of characteristic categories PYC file
Output -- use structural classification (triples, triangles, lines)
Output -- use functional classifcation (triples, triangles, lines)
Output -- use both classifcations (triples, triangles, lines)
Prediction
The frequency-based method (PYC file, result)
The enhanced Frequency-based method (PYC file, result)
The triple method (PYC file, result)
The enhanced triple method (PYC file, result)
Printer Friendly Version 