Research Article

Selecting Negative Samples for PPI Prediction Using Hierarchical Clustering Methodology

Table 4

Description of the 25 extracted features.

Number Description Type

1st # ( 𝐴 𝐺 𝑂 𝐴 𝐵 𝐺 𝑂 𝐴 ) from GOA DB taking 3 ontologies together (P,F,C) Integer
2nd Number of homologs for ( 𝑝 𝑟 𝑜 𝑡 𝐴 , 𝑃 𝑟 𝑜 𝑡 𝐵 ) from HINTdb integer
3rd # [ ( 𝐴 𝑆 𝑃 𝐹 𝐴 𝑀 3 𝐷 𝐼 𝐷 ) + ( 𝐵 𝑆 𝑃 𝐹 𝐴 𝑀 3 𝐷 𝐼 𝐷 ) ] , 𝐴 and 𝐵 are domains extracted form SwissPfam, 3DID is 3did database Integer
4th # ( 𝐴 𝐺 𝑂 𝐴 𝑃 𝐵 𝐺 𝑂 𝐴 𝑃 ) from GOA DB taking Biological Process ontology Integer
5th # ( 𝐴 𝐺 𝑂 𝐴 𝐶 𝐵 𝐺 𝑂 𝐴 𝐶 ) from GOA DB taking Cellular Compartment ontology integer
6th # ( 𝐴 𝐺 𝑂 𝐴 𝐹 𝐵 𝐺 𝑂 𝐴 𝐹 ) from GOA DB taking Molecular Function ontology integer
7th # ( 𝐴 𝑀 𝐼 𝑃 𝑆 𝐹 𝐵 𝑀 𝐼 𝑃 𝑆 𝐹 ) from functional MIPS catalogue identifiers integer
8th # ( 𝐴 𝑀 𝐼 𝑃 𝑆 𝐶 𝐵 𝑀 𝑃 𝐼 𝑆 𝐶 ) from complexes MIPS catalogue identifiers integer
9th # ( 𝐴 𝑀 𝐼 𝑃 𝑆 𝑃 𝐵 𝑀 𝐼 𝑃 𝑆 𝑃 ) from proteins MIPS catalogue identifiers integer
10th # ( 𝐴 𝑀 𝑃 𝐼 𝑆 𝐹 𝐸 𝐵 𝑀 𝑃 𝐼 𝑆 𝐹 𝐸 ) from phenotypes MIPS catalogue identifiers integer
11th # ( 𝐴 𝑀 𝑃 𝐼 𝑆 𝐹 𝐶 𝐶 𝐵 𝑀 𝐼 𝑃 𝑆 𝐹 𝐶 𝐶 ) from subcellular compartments MIPS catalogue identifiers integer
12th Local similarity of 1st feature real
13th Global similarity of 1st feature real
14th # [ ( ( 𝐴 𝑆 𝑃 𝐹 𝐴 𝑀 3 𝐷 𝐼 𝐷 ) + ( 𝐵 𝑆 𝑃 𝐹 𝐴 𝑀 3 𝐷 𝐼 𝐷 ) ) ] / # ( 𝐴 𝑆 𝑃 𝐹 𝐴 𝑀 𝐵 𝑆 𝑃 𝐹 𝐴 𝑀 ) Real
15th Local similarity of 4th feature real
16th Local similarity of 5th feature real
17th Local similarity of 6th feature real
18th Global similarity of 4th feature real
19th Global similarity of 5th feature Real
20th Global similarity of 6th feature Real
21th Local similarity of 7th feature Real
22th Local similarity of 8th feature Real
23th Local similarity of 9th feature Real
24th Local similarity of 10th feature Real
25th Local similarity of 11th feature Real

Symbol # indicates the number of elements in a set. See (2.1) and (2.2).