Inferring putative ancient whole genome duplications in the 1000 Plants (1KP) initiative: access to gene family phylogenies and age distributions

Abstract

Polyploidy or whole genome duplications (WGDs) repeatedly occurred during green plant evolution. To examine the evolutionary history of green plants in a phylogenomic framework, the 1KP project sequenced over 1000 transcriptomes across the Viridiplantae. The 1KP project provided a unique opportunity to study the distribution and occurrence of WGDs across the green plants. In the 1KP capstone analyses, we used a total evidence approach that combined inferences of WGDs from Ks and phylogenomic methods to infer and place ancient WGDs. Overall, 244 putative ancient WGDs were inferred across the Viridiplantae. Here, we describe these analyses and evaluate the consistency of the WGD inferences by comparing them to evidence from published syntenic analyses of plant genome assemblies. We find that our inferences are consistent with whole genome synteny analyses and our total evidence approach may minimize the false positive rate throughout the data set. Given these resources will be useful for many future analyses on gene and genome evolution in green plants, we release 383,679 nuclear gene family phylogenies and 2,306 gene age distribution (Ks) plots from the 1KP capstone paper.

The genome of a cave plant, Primulina huaijiensis, provides insights into adaptation to limestone karst habitats

“…The L event corresponds to known WGD events in other species (Edger et al , 2017; Sollars et al , 2017; Unver et al , 2017; Ren et al , 2018; Li & Barker, 2020). For instance, Sollars et al .…”

“…Nevertheless, they stated that it was unclear if this was two events or just one with skewed signal. Unver et al (2017) and Li & Barker (2020) recognized two lineage‐specific WGDs for O. europaea and F. excelsior , whereas the L event is shared by other Lamiales. We recovered 7524 gene trees of homologs that were used for mapping gene duplication events to the species tree surrounding the L event.…”

Phylogeny and multiple independent whole‐genome duplication events in the Brassicales

“…Recently, an abundance of papers have highlighted the difficulties and complexities of determining WGD events across the tree of life (Conover et al, 2018; Tiley et al, 2018; Li and Barker, 2019 [Preprint]; Li et al, 2019; Nakatani and McLysaght, 2019; Zwaenepoel and van de Peer, 2019 [Preprint]; Zwaenepoel et al, 2019). We add another dimension to this conversation by demonstrating that the different taxonomic levels from which we sample, such as the order or family, make a difference in support of previously identified WGDs (i.e., the Brassiceae triplication).…”

“…We add another dimension to this conversation by demonstrating that the different taxonomic levels from which we sample, such as the order or family, make a difference in support of previously identified WGDs (i.e., the Brassiceae triplication). Recent research has demonstrated that differences in taxonomic sampling and taxon occupancy in data matrices can influence the inference of WGDs, particularly if adding taxa decreases taxon occupancy in gene families (Yang et al, 2015; Li et al, 2018; Li and Barker, 2019 [Preprint]; Zwaenepoel and van de Peer, 2019 [Preprint]). Testing for WGD events across the Brassicales phylogeny led to less certain topologies; therefore, signals of WGD are missed when filtering for nodes with only high bootstrap support to count duplicates.…”

TagSeq for gene expression in non‐model plants: A pilot study at the Santa Rita Experimental Range NEON core site

“…Reference transcriptomes will differ from genomes in that not all loci will be represented by transcripts present in a given sample, not all transcripts will be assembled to full length, and the assembly will vary in the degree to which splice variants, alleles, and paralogs will occur as unique sequences or be merged (Meyer et al, 2011; Yang and Smith, 2013; Carpenter et al, 2019; Patterson et al, 2019). These issues will reduce the number of TagSeq reads that can be uniquely mapped to the reference, relative to a full genome, and they may be particularly problematic in plants where gene and genome duplications are common (Barker et al, 2016; One Thousand Plant Transcriptomes Initiative, 2019; Li and Barker, 2020), although sequencing at the variable 3′ untranslated region (UTR) should maximize locus discrimination (Rise et al, 2004).…”