Variant Identification Workflow for Chromosome Scale Assembly

NGS assemblies of plant genomes often consist of thousands of contigs. Sequencing the segregating progenies is regularly used to anchor the de novo assembled contigs into chromosome-scale assemblies. The analysis of sequencing data from segregating progenies usually involves custom scripting, and requires advanced bioinformatics skills. Here we present a user-friendly workflow that can be performed in CLC Genomics Workbench, enabling biologists to proceed with this type of data analysis.

We used the publicly available ddRAD-Seq data for sacred lotus, Nelumbo nucifera (Liu et al, 2016), and the corresponding genomic assembly consisting of 3,602 contigs. The alignments, as well as all variant calling and variant filtering were performed in CLC Workbench. The variants were called using the Fixed Ploidy Variant caller, filtered against control reads of the other parent, and selected for homozygosity and variant quality. This part of the workflow produced a known variants track, which was used to call variants in the progenies. We further filtered the variant tracks using the CLC Workbench comparative tools, and ended up with 4K variants detected in at least 70% of F2 samples.

To assess the quality of variant calls, we exported the data from CLC Workbench and submitted them to the MadMapper program, which clustered contigs into chromosomes. Most of the genomic assembly (72.5%) was clustered into 9 lotus chromosomes; a similar number, 70.6% was anchored by Liu et al. This confirmed the quality of the marker data for chromosome-scale assemblies outputted by CLC Workbench using the user-friendly workflow.