A unified VCF dataset from nearly 1,500 diverse maize accessions and resources to explore the genomic landscape of maize

Abstract

Efforts to capture and analyze maize nucleotide diversity have ranged widely in scope, but differences in reference genome version and software algorithms used in these efforts inhibit comparison, and these data are generally not available in an easy-to-use visualization platform for quick access and analysis. To address these issues, The Maize Genetics and Genomics Database has collaborated with maize researchers to offer variant data from a diverse set of 1,498 inbred lines, traditional varieties, and teosintes through a standardized variant-calling pipeline against version 5 of the B73 reference genome. The output was filtered for mapping quality, completeness, and linkage disequilibrium, and annotated based on variant effects relative to the B73 RefGen_v5 gene annotations. MaizeGDB has also updated a web tool, SNPversity 2.0, to filter, visualize, and download genotype sets based on genomic locations and accessions of interest, and added external datasets to demonstrate SNPversity 2.0's broad usage. MaizeGDB plans to host annual updates of these resources as additional resequencing data become available, with plans to expand to all publicly available sequence data.

Keywords: diversity; genome; maize; tools; variants.

Fig. 1.

Overview of SNPversity 2.0. a) The “Select options” tab. b) An example of one of the projects, opened to view the accessions within and the option selections. c) The “Table view” output tab featuring a subset of accessions for gene model Zm00001eb56510. d) The “Tree view” output tab from c for gene model Zm00001eb56510.

Fig. 2.

Closer look at the “Select options” tab. a) Overview. b) “Select genomic interval” options. c) “Choose a dataset” options.

Fig. 3.

Example data for the SNPversity 2.0 “Table view” tab. a) Twenty-five NAM founder accessions plus Ab10 for the gene model Zm00001eb407010. b) JBrowse instance of the variant highlighted with the arrow from a. Featured are the reference B73 gene model Zm00001eb407010 at the top, followed by V18 ear RNA-seq track, and the lifted over gene model annotations to genomes corresponding to the NAM founder accessions from a. The accessions from a of CML277 and Tzi8 that have the A−>G stop lost variant are represented by the lifted over gene models from their respective genomes. The A−>G variant from a is represented by the vertical line (arrow). The stop lost variant is associated with extended gene models in CML277 and Tzi8 relative to the other gene models. c) The Tzi8 gene model in its home JBrowse instance. V18 ear RNA-seq expression validates the extended gene model structure relative to B73 and B97.

Fig. 4.

Connecting SNPversity 2.0 to the MaizeGDB PanEffect tool. a) Select the “High-Coverage” dataset from the SNPversity 2.0 “Selection options” tab. b) High-coverage output for gene model Zm00001eb257910. Click on the “missense variant” link in the “Effect type” column to go to the predicted codon missense variant effect prediction in PanEffect. c) The PanEffect visualization platform. By selecting the “Variant effects in B73” tab and the “MaizeGDB 2024 High Coverage variant effects” radio button, the missense variants from the High-Coverage WGS data are shown, with color-coded effect predictions for each missense variant, from blue (benign effect) to red (strong effect). Circled is the P−>L missense variant at codon 296 (P296L) for Zm00001eb257910 from SNPversity 2.0 shown in Fig. 3b.