Figure 4
From: PACIFIC: a lightweight deep-learning classifier of SARS-CoV-2 and co-infecting RNA viruses
False positive and false negative assignments for 100 independently simulated datasets. Left panel—Average number of false positives (y axis). True labels are indicated by colour for each class and predicted labels are given in the x-axis. Right panel—Average number of false negatives (y axis). Predicted labels are indicated by colour for each class and true labels as indicated in the x-axis. pDiscarded—reads that do not reach the 0.95 posterior probability cut-off; rc_discarded—reads that were discordantly predicted using double prediction.
