PhyloHerb: A high-throughput phylogenomic pipeline for processing genome skimming data

Abstract

Premise: The application of high-throughput sequencing, especially to herbarium specimens, is rapidly accelerating biodiversity research. Low-coverage sequencing of total genomic DNA (genome skimming) is particularly promising and can simultaneously recover the plastid, mitochondrial, and nuclear ribosomal regions across hundreds of species. Here, we introduce PhyloHerb, a bioinformatic pipeline to efficiently assemble phylogenomic data sets derived from genome skimming.

Methods and results: PhyloHerb uses either a built-in database or user-specified references to extract orthologous sequences from all three genomes using a BLAST search. It outputs FASTA files and offers a suite of utility functions to assist with alignment, partitioning, concatenation, and phylogeny inference. The program is freely available at https://github.com/lmcai/PhyloHerb/.

Conclusions: We demonstrate that PhyloHerb can accurately identify genes using a published data set from Clusiaceae. We also show via simulations that our approach is effective for highly fragmented assemblies from herbarium specimens and is scalable to thousands of species.

Keywords: herbariomics; high‐throughput sequencing; mitochondria; plastome; ribosomal genes.

Figure 1

PhyloHerb workflow. The five main function modules of PhyloHerb, including qc, getseq, ortho, conc, and order, provide a versatile and efficient tool to curate and analyze genome skimming data.

Figure 2

Defining and extracting genetic blocks with PhyloHerb. (A) A 5‐kbp‐long continuous genetic block on the plastid genome of Arabidopsis thaliana divided into two loci (LOC1 and LOC2). (B) The ‘getseq’ function of PhyloHerb can be used to extract sequences of predefined genetic blocks. The ‘genetic_block’ mode will include genes on both ends, while the ‘intergenic’ mode does not.

Figure 3

Phylogeny of 10 Clusiaceae species inferred from the complete (A) and subsampled plastid data sets (B–D). Raw reads were randomly subsampled to 100 Mbp (B), 50 Mbp (C), and 20 Mbp (D) to simulate decreasing base coverage in genome skimming. For all four analyses, a partitioned concatenated DNA alignment of 87 plastid genes was used to infer the species tree in IQ‐TREE using the GTRGAMMA model. Nodal support was estimated from 1000 ultrafast bootstrap replicates (UFBoot). Unlabeled nodes indicate 100 UFBoot support. Note the unstable placement of Chrysochlamys skutchii in subsampled data sets.