Data from: Horizontal gene transfer fuels metabolic innovation in the grass Zuloagaea bulbosa
Data files
Jul 07, 2026 version files 1.91 GB
-
DatasetS1.tsv
323.93 MB
-
DatasetS2.zip
920.73 KB
-
DatasetS3.zip
1.59 GB
-
README.md
2.57 KB
Abstract
Horizontal gene transfer (HGT) allows the movement of DNA across broad evolutionary distances without sexual reproduction. In grasses, HGT is widespread and although a few horizontally transferred genes (HTG) are adaptive, most are purged over time. Within the adaptive HTG, biosynthetic genes encoding enzymes that act together in the same pathway and physically co-localise in clusters have been reported multiple times. The aims of this study are to test whether HGT is bidirectional in a pair of grass species, maize and Zuloagaea bulbosa, and if HTGs are found more than expected by chance in biosynthetic genes organised in clusters. To achieve this, we firstly generated a phased reference genome for Z. bulbosa. Then we identified 56 candidate horizontally transferred genes, of which 45% were from Andropogoneae, including two likely to be of maize origin. Since transfers from Z. bulbosa to maize were previously described, our results show that HGT is bidirectional, although the balance might not be even. After predicting all biosynthetic gene clusters in the Z. bulbosa genome, we found that HTGs are enriched in biosynthetic genes organised in clusters. This correlation between HGT and gene clustering is likely to be a consequence of selection due to the immediate adaptive benefit a whole pathway can provide. Two of the HTGs from Andropogoneae belong to the benzoxazinoid BGC, which previously underwent an ancestral transfer from Panicoideae into Pooideae. The dynamism of biosynthetic gene clusters, including recurrent horizontal gene transfers, contributes to the extraordinary metabolic diversity present in plants.
Dataset DOI: 10.5061/dryad.tx95x6bd0
Description of the data and file structure
This repository contains the following supplementary data:
Dataset S1. InterProScan functional annotation of Z. bulbosa predicted genes.
Dataset S2. Maximum-likelihood phylogenetic trees for all candidate HTGs, Bx and strigolactone genes.
Dataset S3. Biosynthetic gene cluster prediction from PlantiSmash. The files included in the dataset are the summary spreadsheet containing information about all clusters and summary GenBank file that contains the information for all clusters.
Files and variables
Dataset S1
File DatasetS1.tsv consists of one text file that includes the functional annotations for all the predicted genes in the Zuloagaea bulbosa genome. This information facilitates a rapid interpretation of functional consequences of genomic changes. This dataset may be used by other scientists to predict phenotypic effects of genetic variants, or to expland genomic analyses such as gene family expansions and contractions.
Dataset S2
We used a phylogenetic approach to identify horizontal gene transfer in Z. bulbosa. The gene trees used to evaluate gene tree - species tree discordances, as well as larger phylogenies with expanded sampling for a small subset of the horizontally transferred genes, are included here. Dataset S2 is contained in a compressed file DatasetS2.zip. It includes three subfolders (HTGs, Bx, and Strigolactones). Each subfolder contains the gene trees produced in our work, in newick format. All the details about the methods used to build these trees can be found in the associated pre-print article.
Dataset S3
Some of the horizontally transferred genes were located in biosynthetic gene clusters, so we predicted all the clusters in Z. bulbosa genome. Dataset S3 is contained in a compressed file DatasetS3.zip and contains all the information about the biosynthetic gene clusters in Zuloagaea bulbosa genome predicted by the software PlantiSmash 2.0. This compressed folder contains two files: a spreadsheet with information about all clusters, and a GenBank file with all the details for each cluster, including sequences. This file can be used for visualisation and downstream analyses.
Code/software
All the code and software used to generate these datasets is detailed in the associated pre-print article and in the corresponding GitHub repository.
