Skip to main content
. 2019 Mar 13;28(13):2224–2236. doi: 10.1093/hmg/ddz043

Figure 4.

Figure 4

The PAX7 target gene repression biomarker can discriminate DUX4 target gene-negative FSHD myocytes from controls. (A) A ROC curve compares the discriminatory power of our PAX7 biomarker with the DUX4 target gene signatures of Yao et al. (2014) (30), Geng et al. (2012) (24) and Choi et al. (2016) (34), across just the single cells that showed no expression of the Yao et al. (2014) (30) 114 gene DUX4 signature, as sequenced from two FSHD1, two FSHD2 and two control individuals, published by van den Heuvel et al. (2019) (32). DeLong’s test P-value (P < 2.2 × 10−16) demonstrates that the PAX7 biomarker (AUC = 0.656) is a significantly better discriminator of FSHD status than each of the DUX4 biomarkers (AUC < 0.533). (B) A ROC curve compares the discriminatory power of our PAX7 biomarker with the DUX4 target gene signatures of Yao et al. (2014) (30), Geng et al. (2012) (24) and Choi et al. (2016) (34), across single cells that expressed the Yao et al. (2014) (30) DUX4 signature, as sequenced from two FSHD1, two FSHD2 and two control individuals, published by van den Heuvel et al. (2019) (32). DeLong’s test P-value (P = 0.99) demonstrates that the PAX7 biomarker (AUC = 0.660) is an equivalent discriminator of FSHD status to the Yao et al. (2014) (30) DUX4 signature (AUC = 0.660) on this pre-selected DUX4 target gene expressing population, but superior even to our other two DUX4 signatures (DeLong’s P < 0.0016).