Fig. 3: Validation of HARD RBPs using the protein microarray.
From: Capture of RNA-binding proteins across mouse tissues using HARD-AP
a Schematic of generating a pool of Cy5-labeled RNAs and hybridizing this pool with the human protein microarray. b Scatter plot showing the correlation of the fold change (Cy5 foreground signals over local background) of the HARD-AP identified RBPs (n = 1447) in HEK293 cells. The fold change was calculated from the average of four independent protein spots on the two independent protein arrays (the same for c). The Pearson correlation coefficient is given. c Distributions of the fold change and signal-noise-ratio (SNR) of indicated sets of proteins on the protein microarray (GO RBP n = 1365; HARD-AP RBP n = 1447; HARD-AP specific RBP n = 643). The buffer (n = 320), BSA (n = 80), GST (n = 320), Ig A/G (n = 720) at different concentrations were used as negative controls, and the Alexa 647 labeled IgG (n = 80) was used as the positive control. d Pie chart showing the distribution of fold change (Cy5 signals over local background) of the HARD-AP identified RBPs (n = 1447) and HARD-AP specific RBPs (n = 643) in HEK293 cells available on the protein microarray. e Top GO terms (molecular function and biological process) over-represent in HARD-AP RBPs with FC = 1-1.5 (n = 745) or FC ≥ 1.5 (n = 620) on two independent protein microarrays. The GO enrichment analysis used the two-sided Fisher’s exact test with the p-value adjusted using the Bonferroni correction for multiple testing. The number of proteins is labeled on each bubble. The color scale indicates the enrichment. f Images of Cy5-RNA incubation signal of selected proteins on the protein microarray. g Fold change of subunits of the Integrator, Exosome, Mediator, and 26S proteasome complex. Subunits with fold change over 1.5 were selected for plotting. Data are from four protein spots on two independent protein arrays and shown as the Mean ± SD. Source data for (b–d, g) are provided as a Source Data file.
