Skip to main content
Fig. 4 | BMC Genomics

Fig. 4

From: A machine learning approach for accurate and real-time DNA sequence identification

Fig. 4

Performance analysis of baseline classifiers with respect to (w.r.t) β, Nbins, and H. (a), (b) Accuracy of baseline classifiers w.r.t the R2 test threshold parameter, β, corresponding to target labeling scheme TLS-1 with 6 classes (a) and TLS-2 with 8 classes (b). We chose the same color but different line types to distinguish between datasets [S2,S8], [S6,S9], and [S7], which are of the same strand but use different bias voltages. (c), (d) Accuracy of baseline classifiers w.r.t the number of histogram bins, Nbins, corresponding to target labeling scheme TLS-1 with 6 classes (c) and TLS-2 with 8 classes (d). Similar color scheme and line types as in (a, b) have been used to distinguish between datasets [S2,S8], [S6,S9], and [S7]. (e), (f) Accuracy of baseline classifiers w.r.t the number of traces used to compute a conductance histogram, H, corresponding to target labeling scheme TLS-1 with 6 classes (e) and TLS-2 with 8 classes (bs). We chose the same color but different line types to distinguish between datasets [S2,S8], [S6,S9], and S7, which are of the same strand but use different bias voltages.

Back to article page