Fig. 6: Analysis of clinical samples. | Nature Communications

Fig. 6: Analysis of clinical samples.

From: Empowering the on-site detection of nucleic acids by integrating CRISPR and digital signal processing

Fig. 6

a Cervical brushing samples (n = 121) were analyzed via CreDiT. When compared with clinical diagnoses (bottom row), the CreDiT results showed a good concordance. The CreDiT heatmap displays mean values from triplicate measurements. ∆CreDiT is a background-subtracted signal. b Samples were categorized according to the clinical diagnoses of three major hrHPV targets (HPV16, HPV18, and HPV45). CreDiT signals were significantly higher in clinically positive samples. The p values were <10–6 for all three targets from the unpaired two-sided t-test. Data were displayed as mean ± s.d. c Receiver operating characteristic (ROC) curves were generated for HPV16, HPV18, and HPV45. The predictor was a marker-specific CreDiT signal scaled to the GAPDH signal, considering varying cellularity in a sample. Overall, CreDiT achieved high diagnostic accuracy, with the area under the curve (AUC) values >0.99 for all three HPV types. d CreDiT was used to analyze anal swap samples (n = 48) for anal cancer screening. The results matched the clinical diagnoses (bottom row). The CreDiT heatmap displays mean values from triplicate measurements. e ROC curves were constructed using the GAPDH-scaled marker expression. The AUC values were >0.98. a.u. arbitrary unit. Source data are provided as a Source Data file.

Back to article page