To ensure interactive visualization and seamless operation of genome-wide summary statistics in the web environment, we designed an optimized web architecture to reduce intensive network data transmission load and developed dynamic Manhattan and LocusZoom-like plots. We first introduced a causal block viewer that integrates QQ, Manhattan, and LocusZoom-like plots along with a table displaying credible set variants and a variant annotation panel into a single user-friendly web page (Figure 3). Specifically, QQ (Figure 3A) and Manhattan (Figure 3B) plots can be used to evaluate the quality of GWAS globally. By clicking the highlighted strip or ‘Prev Block’ and ‘Next Block’ buttons in the Manhattan plot, users can switch the causal block of interest. All GWAS variants in each causal block can be displayed in the LocusZoom-like plot (Figure 3C), and users can zoom in or zoom out smoothly using the buttons. Moreover, we added some glyphs to signify the credible set and leading variant. Triangles represent the variants in the credible set and the diamond represents the potential causal variant with leading GWAS signal. When the leading variant is not
