Skip to main content

Table 1 Coefficients of network metrics in two multiple regression models

From: DeepNetBim: deep learning model for predicting HLA-epitope interactions based on network analysis by harnessing binding and immunogenicity information

Binding (linear regression)

HLA attributes

Variable

\({\alpha }_{b}\)

\({\beta }_{HL{A}_{degree}}\)

\({\beta }_{HL{A}_{closeness}}\)

\({\beta }_{HL{A}_{betweeness}}\)

\({\beta }_{HL{A}_{eigenvector}}\)

coefficient

0.281

− 0.048

0.095

0.02

0.003

Peptide attributes

Variable

 

\({\beta }_{PE{P}_{degree}}\)

\({\beta }_{PE{P}_{closeness}}\)

\({\beta }_{PE{P}_{betweeness}}\)

\({\beta }_{{PEP}_{eigenvector}}\)

Coefficient

 

− 0.044

− 0.173

− 0.003

0.05

Immunogenic (logistic regression)

HLA attributes

Variable

\({\alpha }_{i}\)

\({\gamma }_{HL{A}_{degree}}\)

\({\gamma }_{HL{A}_{closeness}}\)

\({\gamma }_{H{LA}_{betweeness}}\)

\({\gamma }_{HL{A}_{eigenvector}}\)

Coefficient

1.914

5.241

− 1.878

− 4.083

− 2.242

Peptide attributes

Variable

 

\({\gamma }_{PE{P}_{degree}}\)

\({\gamma }_{PE{P}_{closeness}}\)

\({\gamma }_{PE{P}_{betweeness}}\)

\({\gamma }_{{PEP}_{eigenvector}}\)

Coefficient

 

− 0.429

3.791

0.092

− 1.022