Extended Data Fig. 9: Receptor-centric isoform distribution in cell lines and single cells.

a, Analysis pipeline combining isoform-level data extracted from transcriptomics experiments of 11 human cell lines deposited in BioProject, with GPCRdb, Ensembl and Uniprot annotations to filter for highly expressed, protein-coding isoforms. Isoforms with truncated signal peptides or those without at least one conserved transmembrane helix as compared with their GPCRdb reference are not considered (see Methods). b, Relationship between the number of GPCRs expressed in the 11 different cell lines (left axis, grey line) and the mean number of isoforms per receptor found in each tissue (right axis; means are shown as black dots and standard errors are grey lines). c, Relationship between the total number of isoforms per receptor and its number of cell-line expression signatures in the analysed cell lines. The dark grey regression line was obtained using a linear model; the light grey confidence interval represents the standard error. d, Mean number of isoforms per receptor expressed in single pancreatic cells as measured by single-cell RNA-seq, with each point representing the mean number of isoforms per receptor in a single cell whose transcriptome was sequenced. This analysis shows that, for every inferred cell type, there are several cells that express more than one isoform per receptor. See underlying data for a–d in Supplementary Table 6. Boxplots show median and 25th and 75th percentiles; whiskers represent 1.5 times the interquartile range from the 25th and 75th quartiles.