Skip to main content

Advertisement

Fig. 3 | Journal of Translational Medicine

Fig. 3

From: Development of an objective gene expression panel as an alternative to self-reported symptom scores in human influenza challenge trials

Fig. 3

Intensities from Illumina ht-12 arrays of the 19 gene signature were used to train a random forest model. The model was used to predict symptom scores in an independent test set. a Actual scores were plotted against the predicted values at each day. RMSE decrease and correlation values refer to the overall model fit. b Importance of individual markers within the panel of 19 primers. X-axis shows the increase in RMSE if the respective gene is left out of the model

Back to article page