Figure 2.

In silico analysis identifies splicing factors potentially regulating the generation of MERS-CoV-perturbed circRNA-mRNA pairs. (A) Heatmap presentation of MERS-CoV-perturbed differentially expressed (DE) splicing factors. The splicing factors were under hierarchical clustering and coloured by its normalized intensity scale [log₂ (expression values in transcripts per million)]. (B) Candidate splicing factors potentially governing the expression of the representative circRNA-cognate mRNA pairs perturbed in MERS-CoV infection. The predicted binding motifs between each circRNA and splicing factor, Pearson correlation coefficients between each circRNA and splicing factor (PCC_circ) with its corresponding correlation P value (cP_circ), and Pearson correlation coefficients between each mRNA and splicing factor (PCC_m) with its corresponding correlation P value (cP_m) were listed. (C) Network demonstrating the interactions among MERS-CoV-perturbed circRNAs and splicing factors. Splicing factors and circRNAs were represented by rectangular and circular nodes, respectively. The general expression trend of each splicing factor during MERS-CoV infection was distinguished by filling colour (orange: up-regulated; blue: down-regulated). The size of splicing factors was proportional to the number of circRNAs potentially under their regulation. Among the seven splicing factors, HNRNPC and U2AF2 stood out to be the largest ones as they were predicted to regulate the highest number of circRNAs (10 circRNAs) over other candidate splicing factors. The filling colour density of each circRNA was mapped according to the number of its potential interactive splicing factors. Edge colour and thickness were proportionally correlated with the correlation coefficiency and correlation P value between each circRNA and splicing factor, respectively.