SNiPlay: a web-based tool for detection, management and analysis of SNPs. Application to grapevine diversity projects

Abstract

Background: High-throughput re-sequencing, new genotyping technologies and the availability of reference genomes allow the extensive characterization of Single Nucleotide Polymorphisms (SNPs) and insertion/deletion events (indels) in many plant species. The rapidly increasing amount of re-sequencing and genotyping data generated by large-scale genetic diversity projects requires the development of integrated bioinformatics tools able to efficiently manage, analyze, and combine these genetic data with genome structure and external data.

Results: In this context, we developed SNiPlay, a flexible, user-friendly and integrative web-based tool dedicated to polymorphism discovery and analysis. It integrates:1) a pipeline, freely accessible through the internet, combining existing softwares with new tools to detect SNPs and to compute different types of statistical indices and graphical layouts for SNP data. From standard sequence alignments, genotyping data or Sanger sequencing traces given as input, SNiPlay detects SNPs and indels events and outputs submission files for the design of Illumina's SNP chips. Subsequently, it sends sequences and genotyping data into a series of modules in charge of various processes: physical mapping to a reference genome, annotation (genomic position, intron/exon location, synonymous/non-synonymous substitutions), SNP frequency determination in user-defined groups, haplotype reconstruction and network, linkage disequilibrium evaluation, and diversity analysis (Pi, Watterson's Theta, Tajima's D).Furthermore, the pipeline allows the use of external data (such as phenotype, geographic origin, taxa, stratification) to define groups and compare statistical indices.2) a database storing polymorphisms, genotyping data and grapevine sequences released by public and private projects. It allows the user to retrieve SNPs using various filters (such as genomic position, missing data, polymorphism type, allele frequency), to compare SNP patterns between populations, and to export genotyping data or sequences in various formats.

Conclusions: Our experiments on grapevine genetic projects showed that SNiPlay allows geneticists to rapidly obtain advanced results in several key research areas of plant genetic diversity. Both the management and treatment of large amounts of SNP data are rendered considerably easier for end-users through automation and integration. Current developments are taking into account new advances in high-throughput technologies.SNiPlay is available at: http://sniplay.cirad.fr/.

Figure 1

SNiPlay overview. This figure illustrates the analysis pipeline implemented in SNiPlay and its relationship with the database. The workflow consists of seven steps: SNP/indel detection, Mapping, Functional annotation, Phasing, Linkage Disequilibrium, Diversity analysis, Haplotype network. The database has been augmented after launching the first two modules of the pipeline.

Figure 2

Overview of the pipeline process and graphical outputs. (A) Input can be in the form of electrophoregrams, FASTA multiple alignments or genotyping data. (B) Accession selection. (C) Different types of outputs and graphical layouts.

Figure 3

Web interface for SNP queries. The system is able to identify SNPs in a user-defined subset of genotypes (A) considering a set of selected genes (B). It can merge allelic data from different origins and different experiments and report polymorphic positions in different colors (C): green if no difference appeared when merging data, yellow if some differences were detected and white if no merging is needed.

Figure 4

Examples of global statistics for a given project. Graphical outputs for a grapevine project (A) Mean heterozygosity of 47 grapevine accessions over 1,111 amplicons (number of heterozygous position over a 10,000 bp region). (B) Frequency of alternative SNP variations. (C) Distribution of indel sizes.

Figure 5

Diversity map. Tajima's values are color-coded along the chromosomes. A normal distribution of Tajima's values is assumed to estimate significance: significantly negative values (in red; p < 0.05) suggest an excess of low frequency SNPs, indicating putative regions under purifying selection, while high values (in green) indicate regions where diversification probably occurred.

Figure 6

Haplotype blocks and LD visualization with Haploview. for each genomic region submitted to SNiPlay, LD scores are calculated for each SNP pairs using the phased genotypes and reported in a LD plot generated by Haploview (via the Gevalt program).

Figure 7

Venn diagrams. SNiPlay reports 2 kinds of Venn diagrams indicating (A) the number of polymorphisms (SNPs + indels) shared between groups and (B) the cumulative number of polymorphisms when combining groups.

Figure 8

Example of comparative diversity maps. (A) A comparison of Tajima's D values over chromosomes 17 and 18 between cultivated and wild grapevine. An asterisk indicates genes for which the difference in Tajima's D value between groups is significant. This map comparison is possible only between the first 2 groups. (B) A genomic region near a "berry size" QTL displays differential D values, which are being further investigated to test for potential association with the wild-cultivated berry size differential.

Figure 9

Example of a haplotype network. The sizes of the circles are proportional to haplotype frequency, and the lengths of the connecting lines are proportional to the number of mutational steps between haplotypes. External information associated with the genotypes (south, east and west) was specified by the user so that the circles are replaced by colored pies showing the frequency of each subgroup.