SNP genotyping arrays have been useful for many applications that require a large number of molecular markers such as high-density genetic mapping, genome-wide association studies (GWAS), and genomic selection. We report the establishment of a large maize SNP array and its use for diversity analysis and high density linkage mapping. The markers, taken from more than 800,000 SNPs, were selected to be preferentially located in genes and evenly distributed across the genome. The array was tested with a set of maize germplasm including North American and European inbred lines, parent/F1 combinations, and distantly related teosinte material. A total of 49,585 markers, including 33,417 within 17,520 different genes and 16,168 outside genes, were of good quality for genotyping, with an average failure rate of 4% and rates up to 8% in specific germplasm. To demonstrate this array's use in genetic mapping and for the independent validation of the B73 sequence assembly, two intermated maize recombinant inbred line populations – IBM (B73×Mo17) and LHRF (F2×F252) – were genotyped to establish two high density linkage maps with 20,913 and 14,524 markers respectively. 172 mapped markers were absent in the current B73 assembly and their placement can be used for future improvements of the B73 reference sequence. Colinearity of the genetic and physical maps was mostly conserved with some exceptions that suggest errors in the B73 assembly. Five major regions containing non-colinearities were identified on chromosomes 2, 3, 6, 7 and 9, and are supported by both independent genetic maps. Four additional non-colinear regions were found on the LHRF map only; they may be due to a lower density of IBM markers in those regions or to true structural rearrangements between lines. Given the array's high quality, it will be a valuable resource for maize genetics and many aspects of maize breeding.

The topology of cellular circuits (the who-interacts-with-whom) is key to understand their robustness to both mutations and noise. The reason is that many biochemical parameters driving circuit behavior vary extensively and are thus not fine-tuned. Existing work in this area asks to what extent the function of any one given circuit is robust. But is high robustness truly remarkable, or would it be expected for many circuits of similar topology? And how can high robustness come about through gradual Darwinian evolution that changes circuit topology gradually, one interaction at a time? We here ask these questions for a model of transcriptional regulation networks, in which we explore millions of different network topologies. Robustness to mutations and noise are correlated in these networks. They show a skewed distribution, with a very small number of networks being vastly more robust than the rest. All networks that attain a given gene expression state can be organized into a graph whose nodes are networks that differ in their topology. Remarkably, this graph is connected and can be easily traversed by gradual changes of network topologies. Thus, robustness is an evolvable property. This connectedness and evolvability of robust networks may be a general organizational principle of biological networks. In addition, it exists also for RNA and protein structures, and may thus be a general organizational principle of all biological systems. Living things are astonishingly complex, yet unlike houses of cards they are also highly robust. That is, they have persisted for billions of years, despite being exposed to an endless stream of environmental stressors and random mutations. Is this robustness an evolvable property? Do different biological systems vary in their robustness? Has natural selection shaped this robustness? These questions are very difficult to answer experimentally for most systems, be they proteins or large gene networks. Here we address these questions with a model of the transcription regulation networks that regulate both cellular functions and embryonic development in many organisms. We examine millions of such networks that differ in the topology or architecture of their regulatory interactions, that is, in the “who interacts with whom” of a network. We find that radically different network architectures can show the same gene expression pattern. The networks' robustness to both mutations and gene expression noise shows a broad distribution: some network architectures are highly robust, whereas others are quite fragile. Importantly, the entire space of network architectures can be traversed through small changes of individual regulatory interactions, without changing a network's gene expression pattern. This means that high robustness in gene expression can evolve through gradual and neutral evolution in the space of network architectures. Our results show that the robustness of transcriptional regulation networks is an evolvable trait that natural selection can change like any other trait.

The history of life involves countless evolutionary innovations, a steady stream of ingenuity that has been flowing for more than 3 billion years. Very little is known about the principles of biological organization that allow such innovation. Here, we examine these principles for evolutionary innovation in gene expression patterns. To this end, we study a model for the transcriptional regulation networks that are at the heart of embryonic development. A genotype corresponds to a regulatory network of a given topology, and a phenotype corresponds to a steady-state gene expression pattern. Networks with the same phenotype form a connected graph in genotype space, where two networks are immediate neighbors if they differ by one regulatory interaction. We show that an evolutionary search on this graph can reach genotypes that are as different from each other as if they were chosen at random in genotype space, allowing evolutionary access to different kinds of innovation while staying close to a viable phenotype. Thus, although robustness to mutations may hinder innovation in the short term, we conclude that long-term innovation in gene expression patterns can only emerge in the presence of the robustness caused by connected genotype graphs.

In most species, crossovers (COs) are essential for the accurate segregation of homologous chromosomes at the first meiotic division. Their number and location are tightly regulated. Here, we report a detailed, genome-wide characterization of the rate and localization of COs in Arabidopsis thaliana, in male and female meiosis. We observed dramatic differences between male and female meiosis which included: (i) genetic map length; 575 cM versus 332 cM respectively; (ii) CO distribution patterns: male CO rates were very high at both ends of each chromosome, whereas female CO rates were very low; (iii) correlations between CO rates and various chromosome features: female CO rates correlated strongly and negatively with GC content and gene density but positively with transposable elements (TEs) density, whereas male CO rates correlated positively with the CpG ratio. However, except for CpG, the correlations could be explained by the unequal repartition of these sequences along the Arabidopsis chromosome. For both male and female meiosis, the number of COs per chromosome correlates with chromosome size expressed either in base pairs or as synaptonemal complex length. Finally, we show that interference modulates the CO distribution both in male and female meiosis. Reciprocal exchanges of genetic material (crossovers) between homologous chromosomes ensure their proper segregation to generate gametes. Their number and location along chromosomes are tightly regulated. We localized precisely the position of 13,535 crossovers in more than 3,000 plants of Arabidopsis thaliana. While A. thaliana is a hermaphrodite plant with male and female meiosis occurring in the same flower and thus with the same genome, we observed dramatic differences in the distribution and the rate of crossovers along chromosomes in male and female meiosis. On average, chromosomes recombine 1.7 times more in male than in female meiosis. Moreover, male CO rates are very high at both ends of each chromosome, whereas female CO rates are very low. Finally, for the first time in a eukaryote, we show that the correlations between CO rates and various chromosome features differ in male and female meiosis. Female CO rates correlated strongly and negatively with GC content and gene density but positively with transposable elements density, whereas male CO rates correlated positively with the CpG ratio. However, most of the correlations could be explained by the structure of the Arabidopsis genome.