Skip to main content

Table 4 Parameter used to optimize each network-inference algorithms

From: A machine learning approach for the identification of key markers involved in brain development from single-cell transcriptomic data

Network-inference algorithms

Parameter optimization settingc

GENIE3-Aa

K = “all”, nb.trees = 10,000

GENIE3-Ba

K = “sqrt”, nb.trees = 10,000

TIGRESS-Ab

scoring = “area”

TIGRESS-Bb

scoring = “max”

ARACNE

eps = 0.1

BC3NET

boot = 10, alpha1 = 0.99, alpha2 = 0.99

SiGN-BN

Number of iteration of bootstrap method = 1,000

  1. aGENIE3-A and -B represent two different parameter settings for GENIE3 algorithm used in this study
  2. bTIGRESS-A and -B represent two different parameter settings for TIGRESS algorithm used in this study
  3. cWe used default settings for parameters that are not shown in this table