Automated assembly scaffolding using RagTag elevates a new tomato system for high-throughput genome editing

Abstract

Advancing crop genomics requires efficient genetic systems enabled by high-quality personalized genome assemblies. Here, we introduce RagTag, a toolset for automating assembly scaffolding and patching, and we establish chromosome-scale reference genomes for the widely used tomato genotype M82 along with Sweet-100, a new rapid-cycling genotype that we developed to accelerate functional genomics and genome editing in tomato. This work outlines strategies to rapidly expand genetic systems and genomic resources in other plant species.

Keywords: Assembly scaffolding; Genome editing; Genome sequencing; Tomato.

Fig. 1

RagTag enables rapid generation of new reference genomes for the tomato genotypes Sweet-100 and M82. a Images of M82, Micro-tom (MT), and Sweet-100 (S100) plants 44 days (top) and 65 days (bottom) after sowing. Red asterisks indicate open flowers. b Number of inflorescences with open flowers (top), green fruits (middle), and ripe fruits (bottom) at 6 to 9 weeks after sowing. Data points represent individual plants (n=8). c Images of the first developing fruit on M82, MT, and S100 at 65 days after sowing. d Diagram indicating generation times of the M82, MT, and S100 genotypes. e Overview of RagTag “scaffold,” “patch,” and “merge.” f A more detailed diagram describing RagTag “patch”, highlighting how sequence from the query assembly (orange) can be used to fill gaps in the target assembly (green). g A more detailed diagram describing RagTag “merge” showing how each contig is represented by a pair of nodes for the beginning and end termini of the sequence with edges between contigs indicating the pair of contigs are adjacent in one of the candidate scaffolds. The function h() maps contig terminus pairs to Hi-C scores (see Methods section “RagTag “merge””). h nX plots showing the minimum sequence length (y-axis, log scale) needed to constitute a particular percentage of the assembly (x-axis). i Ideogram showing contig boundaries (alternating color and gray) within the final scaffolds. j Circos plots comparing M82 to Heinz 1706 (SL4.0). Circos quantitative tracks a, b, and c are summed in 500 kbp windows and show number of genes (a, lower tick=0, middle tick=47, upper tick=94), LTR retrotransposons (b, 0, 237, 474) and structural variants (c, 0, 24, 48). The inner ribbon track shows whole-genome alignments, with blue indicating forward-strand alignments and red indicating reverse-strand alignments (inversions). Darker colors indicate alignment boundaries. k, As for j but comparing Sweet-100 to Heinz 1706 and showing number of genes (a, 0, 48, 96), LTR retrotransposons (b, 0, 269, 538), and structural variants (c, 0, 30, 59) and whole-genome alignment ribbons. Letters in b represent post hoc Tukey’s HSD tests. Scale bars indicate 10 cm (a) and 1 cm (c)

Fig. 2

Sweet-100 is an effective system for genome editing experiments. a CRISPR-Cas9 targeting of SlAP3 using two gRNAs. Black boxes, black lines, and blue boxes represent exonic, intronic, and untranslated regions, respectively. b Images of detached inflorescences (top) and flowers (bottom) from wild-type (WT) and seven independent first-generation (T0) ap3^CR transgenic plants. c and d CRISPR-induced mutations in SlAP3 identified by agarose gels (c) and Sanger sequencing (d). gRNA and PAM sequences are indicated in red and black bold letters, respectively; deletions are indicated with blue dashes; sequence gap length is given in parenthesis. e Full gene deletion of AN by CRISPR-Cas9 using two gRNAs. f and g, Detection of complete deletion of the AN gene by agarose gel electrophoresis (f) and Sanger sequencing (g). h Images of WT and an^CR mutant plants in the non-transgenic second (F2) generation. i CRISPR-Cas9 targeting of the SEP4 gene family using five gRNAs. j analysis of j2^CRej2^CRlin^CR T0 plants by agarose gel electrophoresis, k, images of T0 plants showing j2^CRej2^CR double (T0–6) and j2^CRej2^CRlin^CR triple (T0–17) mutant phenotypes. l High-throughput discovery of CRISPR-Cas9 mutations in J2, EJ2, and LIN by multiplexed amplicon sequencing. Heatmap shows the percentage of modified reads in 184 T1 and 8 WT control plants. Red font indicates WT control individuals. Dotplot depicts the number of branches on 1 to 5 inflorescences per individual plant. m Percentage of modified reads in WT, all T1, and individual T1 families; n equals the number of individual plants. n Percentage of inflorescences with 1 to 5 or more branches on plants in (m); n and N equal the number of individual plants and inflorescences, respectively. o Sequences and frequency of edited alleles identified from in the T1 generation. p Images of detached inflorescences from individual plants from (l); percentage of modified reads (%) and number of inflorescence branches (mean ± s.d.) are indicated. Scale bars indicate 1 cm