Skip to main content
Figure 2 | Human Genomics

Figure 2

From: A general integrative genomic feature transcription factor binding site prediction method applied to analysis of USF1 binding in cardiovascular disease

Figure 2

Receiver operator characteristic curve (ROC) of USF1 prediction models. The curves generated by the SPSS package [27] with different colours indicate the sensitivity and specificity of three different prediction models. The sensitivity and specificity were calculated from LOLO cross-validation (see Methods). Sensitivity was the proportion of correctly predicted positive loci whereas specificity was the proportion of correctly predicted negative loci. The all features model used all five genomic features (PhastCons8, MostCons8, RP5, CpG and DNaseI HS) and PWM score, and the selected features model only included three genomic features (PhastCons8, RP5 and DNaseI HS) and PWM score.

Back to article page