Skip to main content

Table 2 Performance of host tropism predictive models for individual proteins on independent training and testing data

From: HopPER: an adaptive model for probability estimation of influenza reassortment through host prediction

Model

Training data

Testing data

 

Accuracy

Precision

Recall

G-means

MCC

Accuracy

Precision

Recall

G-means

MCC

HA

0.966

0.967

0.956

0.953

0.943

0.965

0.969

0.956

0.955

0.947

NA

0.961

0.962

0.953

0.953

0.939

0.957

0.958

0.95

0.949

0.933

NP

0.947

0.944

0.933

0.931

0.912

0.954

0.951

0.944

0.943

0.927

PA

0.929

0.916

0.893

0.89

0.881

0.922

0.906

0.892

0.888

0.875

PB1

0.931

0.927

0.907

0.902

0.887

0.937

0.933

0.914

0.912

0.898

PB2

0.943

0.937

0.912

0.913

0.906

0.945

0.938

0.923

0.921

0.911

NS1

0.934

0.928

0.917

0.916

0.896

0.931

0.93

0.919

0.917

0.896

M2

0.876

0.866

0.856

0.854

0.805

0.865

0.86

0.853

0.848

0.795