Skip to main content

Table 7 Comparison between the degree distribution properties of the yeast interactome by source database.

From: iRefR: an R package to manipulate the iRefIndex consolidated protein interaction database

database

n

dmax

dmin

alpha

ntail

p-value

power-law

DIP

5189

351

79

3.46

177

0.04

no

BIND

5124

316

23

3.25

292

0.84

yes

IntAct

5893

969

55

3.03

439

0.26

yes

MINT

5523

390

33

3.33

310

0.92

yes

BioGrid

5794

1305

239

4.10

194

0.01

no

MPACT

4953

281

13

2.56

711

0.01

no

Total

6875

1327

199

3.63

459

0.00

no

  1. The table shows the degree distribution properties of all seven databases reporting interactions with at least one yeast interactor in the record. The canonical representation is used to normalize protein interactors. N-ary data was represented using a spoke model. Data may include binary records that are representing n-ary data. Results may differ under other conditions.
  2. Half of the databases (BIND, IntAct and MINT) suggest a power-law behaviour for the yeast interactome while the other half and the consolidated network cannot be fitted to a power-law.
  3. n: number of distinct protein interactors in network
  4. dmax: maximum node degree in network
  5. dmin: minimum node degree considered for power-law fitting
  6. alpha: fitted power-law exponent
  7. ntail: number of nodes in tail of distribution considered for power-law fitting
  8. p-value: power-law fitness- > 0.1 guarantees that the data can be fitted to a power-law model, however, better fits may exist for other functions