Analysis of INM hits shows overlap with the ER. (A) A scoring matrix for INM hits based on detection at the INM (binned into four categories based on laser power) and whole-cell protein abundance data from Ghaemmaghami et al. (2003) (binned into four categories). In this way, 242 of 412 hits were ranked: blue and dark green, high-confidence hits; light green, medium confidence; yellow, orange, or red, low confidence. Laser power, abundance, and confidence score for each INM hit listed in Table S4. No abundance data were available for 170 hits, so they were not scored, although most are likely to be high-confidence hits because protein detection fell below detection limits in previous work. Because of the use of the CDC42pr, some genes in this category may represent products not normally expressed under standard vegetative growth conditions. (B) GO slim component analysis of the proteins in each confidence category of low, medium, and high. (C) Enrichment for cellular components based on GO slim annotations in SGD was performed for the library (left) and for genes in each category (right). (D) For single-pass integral membrane proteins localized to the INM (n = 65), sizes of cytoplasmic/nucleoplasmic and luminal domains were determined based on amino acid composition. Total molecular masses for INM hits as well as single-pass proteins that were negative (n = 110)/negative with GO component term: endomembrane system (n = 54) were plotted. The mean and SEM are shown on each plot. P-values were calculated by Student’s t test. One split-GFP negative was removed by the Grubs test: Csf1, molecular mass = 338 kD.