Phylogenetics benefits from large data sets of single-copy nuclear genes (SCG) obtainable from next-generation sequencing (NGS) data, which has revolutionized plant phylogenetics. Target enrichment or sequence capture is a DNA-based, efficient and cost-effective method that uses hybridization of sheared whole genomic DNAs from the species of interest to conserved sequence oligonucleotides. Combining target enrichment with NGS strategies has the potential to yield a large number of low-copy nuclear loci and is becoming progressively popular for phylogenomics. Depending on availability and quality of reference data, (semi-)automated identification of suitable markers will require specific and fine-tuned bioinformatic pipelines. Here we establish a highly efficient bioinformatic pipeline to discover putative orthologous SCGs from transcriptome data of the plant family Orobanchaceae. This family contains species with contrasting nutritional modes including parasitic species, some of which are pests on crop plants. We used transcriptome data of differing quality available for four Orobanchaceae species, using Mimulus guttatus (1,922 seqs) and Solanum lycopersicum SCGs (374 seqs) as SCG reference datasets. Depending on whether gaps were permitted in initial BLAST searches of the four Orobanchaceae species against the reference genomes or not, our pipeline identified 1195 and 993 SCGs, respectively, of length of at least 780 bps. Automated bait sequence construction (using 2× tiling) resulted in c. 38,000 and 24,000, respectively, bait sequences. We compared our pipeline to the recently published MarkerMiner 1.0 pipeline, but MarkerMiner identified only 533 SCGs. Our pipeline is suitable for other non-model plants, where transcriptomes or partial genome data are available.
