Skip to main content
Fig. 4 | Genome Medicine

Fig. 4

From: A machine learning classifier using 33 host immune response mRNAs accurately distinguishes viral and non-viral acute respiratory illnesses in nasal swab samples

Fig. 4

Performance of LOGR model of the 33 mRNAs in training and validation datasets. Datasets were split the same way as in original discovery (Table 1). Logistic Regression model was trained using expression values of the 33 mRNAs in 6 discovery datasets (A) achieving summary AUC = 0.94 and then applied as a locked model to 4 validation datasets without co-normalization (B) achieving summary AUC of 0.89. In validation, GSE163151 was split into two discovery datasets to represent healthy and non-viral ARI separately

Back to article page

Genome Medicine

ISSN: 1756-994X

Contact us