Fig. 1

Protein-coding genes lost during vertebrate evolution. a Phylogeny of the species used for detection of “genes with lost coding potential” (GLCPs) and the numbers of GLCPs found in each of the reference species. b Fraction of GLCPs and other genes that belong to Ensembl protein families with more than one member. P values were computed using Pearson’s chi-squared test. c Tissue specificity indices [51] of the indicated groups of protein-coding genes in each of the reference species: (i) GLCPs belonging to Ensembl protein families with multiple members (GLCP multi-family); (ii) of other members of the same protein families (Other multi-family; the number of members from each family sampled to be as close as possible to the number of GLCPs in the family); (iii) of GLCPs that belong to families without additional members (GLCP singleton); and (iv) of other genes that belong to families without additional members (Other singleton). Numbers indicate the size of each group. All comparisons indicated by asterisks are significant at FDR < 0.05 (Benjamini–Hochberg method). d Genomic arrangements of genes in the three syntenic genomic clusters surrounding LNX genes in chicken and human. The shaded region highlights the genes in the X inactivation center (XIC), the GLCPs they were derived from, and their paralogs. Gene positions taken from the UCSC genome browser. For genes with multiple splice isoforms, a single representative transcript is shown. Gene model colors indicate the orientation of the gene. Circled numbers indicate assignment of genes to homology groups