Constrained non-coding sequence provides insights into regulatory elements and loss of gene expression in maize

Abstract

DNA sequencing technology has advanced so quickly, identifying key functional regions using evolutionary approaches is required to understand how those genomes work. This research develops a sensitive sequence alignment approach to identify functional constrained non-coding sequences in the Andropogoneae tribe. The grass tribe Andropogoneae contains several crop species descended from a common ancestor ~18 million years ago. Despite broadly similar phenotypes, they have tremendous genomic diversity with a broad range of ploidy levels and transposons. These features make Andropogoneae a powerful system for studying conserved non-coding sequence (CNS), here we used it to understand the function of CNS in maize. We find that 86% of CNS comprise known genomic elements e.g., cis-regulatory elements, chromosome interactions, introns, several transposable element superfamilies, and are linked to genomic regions related to DNA replication initiation, DNA methylation and histone modification. In maize, we show that CNSs regulate gene expression and variants in CNS are associated with phenotypic variance, and rare CNS absence contributes to loss of gene expression. Furthermore, we find the evolution of CNS is associated with the functional diversification of duplicated genes in the context of the maize subgenomes. Our results provide a quantitative understanding of constrained non-coding elements and identify functional non-coding variation in maize.

De novo assembly, annotation, and comparative analysis of 26 diverse maize genomes

“…The phylogeny was visualised and re-rooted on the longest internal branch in iTOL(142).Population Genetic AnalysisGERP. Soft masked copies of 13 angiosperm genomes were aligned to the unmasked B73v5 reference genome using LAST(143)(144)(145)(146)(147)(148). Repetitive elements in B73v5 were then masked in the aligned sequences.…”

Variation in upstream open reading frames contributes to allelic diversity in protein abundance

“…Studies identifying CPuORFs have also been largely focused on dicots. However, many of the uORFs containing derepressive and repressive variants that we identified are conserved between members of the Andropogoneae tribe 41 , although they are not significantly more or less conserved than all other uORFs (Supplemental Figure 3).…”

“…Alignments between maize and five other grasses from the Andropogoneae tribe (45) were filtered to regions that overlap the annotated B73 5' UTRs. uORFs, identified above, were classified based on whether at least one other species had alignments covering 90% or more of the uORF.…”

A multiple genome alignment workflow shows the impact of repeat masking and parameter tuning on alignment of functional regions in plants

“…A drawback is that repeat elements not similar to those in the library will not be masked and, conversely, non-repetitive functional elements with similarities to repeats may be erroneously masked (Bayer, Edwards, and Batley 2018). To compare kmer-based masking approaches to RepeatMasker, we therefore also conducted masking with RED and a novel kmer-based approach (Song et al, 2020) that we refer to as KMER.…”

“…Appropriate repeat masking can improve multiple alignment performance A major obstacle to accurate and efficient alignment is the large proportion of repetitive sequence found in most plant genomes. In contrast to masking tools like RepeatMasker that rely on repeat databases, approaches such as RED (Girgis 2015) or KMER (Song et al 2020) try to avoid database bias by using repetitive k-mers (nucleotide sequences of k length) in the genome to identify repeats. Here, we compared RepeatMasker, RED and KMER and tested their impact on subsequent multiple sequence alignment in grasses.…”