This spreadsheet consists of a catalog of 1008 genetic variants of evolutionary relevance published up to December 2012, and aims at being used as a bibliographical resource for the exploration of evolutionary genetics literature and for detecting emerging patterns in this fast-evolving field. The IL Hotspot column associates three classes of interlineage hotspots to each orthologous group. *** (N=357): alleles of orthologous genes for which genotype-to-phenotype association was verified in a cross (Linkage Mapping) for at least two derived alleles. ** (N=254): alleles of orthologous genes where a derived allele was discovered by Linkage Mapping, or where all the entries are associated to Candidate Gene or Association Mapping approaches. * (N=57): putative hotspot genes associated to ambiguous orthology relationships or phenotypic similarities between lineages, shown here for heuristic purposes. Together, the most conservative interlineage hotspot classes (** and ***) span 111 orthologous groups that encompass 60.6% (611/1008) of the alleles in the dataset. This updated catalog builds on a previous list of 395 mutations (Stern and Orgogozo 2008), and has now expanded to 1008 entries. To identify relevant studies, we used multiple search methods. We screened every issue of the major journals in evolutionary genetics for the last five years (update from (Stern and Orgogozo 2008), performed countless keyword searches on online search tools, and paid particular attention to citations in primary research articles as well as in review papers. The list comprises findings from: 1) Linkage Mapping (top-down) studies that verify genotype-phenotype co-segregation in crosses; 2) Association Mapping (top-down) genome-wide association studies (GWAS) between between heterogeneous individuals from naturally reproducing populations (these are here limited to variants tightly linked to well-supported candidate genes). 3) Candidate Gene (bottom-up) studies that, with various arguments, suggest a causal relationship between genotype and phenotype without co-segregation tests. Importantly, we excluded candidate gene studies that lack evidence of sequence-level change, and for instance, the proposed link between BMP4 and the microevolution of beak shape in Darwin's finches (Campas et al. 2010) was ignored here, as its differential expression between beak morphs could be due to mutations in upstream regulator genes. To take experimental biases into account, we assigned to each evolutionary change one category among these three, that best reflects the type of evidence associating a given allele to a phenotypic effect. We attempted to compile a comprehensive list of linkage mapping studies that identified genetic loci responsible for phenotypic evolution as these studies are initially blind in terms of genotype inferences and avoid spurious effects of gene re-discovery due to ascertainment biases. In contrast, the Candidate Gene and Association Mapping categories are still largely incomplete. With a few exceptions, Candidate gene studies that identified causal but fixed sequence differences between distant species have not been included. This is not only because compiling the relevant articles would require a unreasonable effort, but also because focusing on microevolutionary time scales avoids the confounding effect of cumulative changes over long times of divergence (see Stern 2000 for a discussion of evolutionarily-relevant mutations). We also note that the Candidate Gene category also reflects strong ascertainment biases inherent to this approach, as illustrated by the fact that MC1R alleles make a sixth of all entries in this category (70/413=16.9%). The Association Mapping category was purely the product of cherry picking, but we kept it for comparative and heuristic purposes. It consists of 101 well-supported variants isolated from genome-wide association studies, among which 55 entries concern human studies, 18 were found in domesticated species (mostly in dog), and 18 relate to experimental evolution studies (mostly in yeast). The catalog is thus primarily based on top-down studies, with about half (494/1008) of entries belonging to the Linkage Mapping category. It is also inherently biased towards large-effect QTLs, the low hanging fruits of phenotypic variation (Rockman 2011), and towards shallow phylogenetic levels, because linkage mapping is not feasible in species that do not hybridize. References: Campas, O., R. Mallarino, A. Herrel, A. Abzhanov, and M. P. Brenner. 2010. Scaling and shear transformations capture beak shape variation in Darwin's finches. Proc. Nat. Acad. Sci. USA 107:3356-3360. Rockman, M. V. 2011. The QTN program and the alleles that matter for evolution : All that's gold does not glitter. Evolution 66:1-17. Stern, D. 2000. Evolutionary developmental biology and the problem of variation. Evolution 54:1079-1091. Stern, D. L., and V. Orgogozo. 2008. The loci of evolution: How predictable is genetic evolution? Evolution 62:2155-2177.