The Release 6 reference sequence of the Drosophila melanogaster genome

Abstract

Drosophila melanogaster plays an important role in molecular, genetic, and genomic studies of heredity, development, metabolism, behavior, and human disease. The initial reference genome sequence reported more than a decade ago had a profound impact on progress in Drosophila research, and improving the accuracy and completeness of this sequence continues to be important to further progress. We previously described improvement of the 117-Mb sequence in the euchromatic portion of the genome and 21 Mb in the heterochromatic portion, using a whole-genome shotgun assembly, BAC physical mapping, and clone-based finishing. Here, we report an improved reference sequence of the single-copy and middle-repetitive regions of the genome, produced using cytogenetic mapping to mitotic and polytene chromosomes, clone-based finishing and BAC fingerprint verification, ordering of scaffolds by alignment to cDNA sequences, incorporation of other map and sequence data, and validation by whole-genome optical restriction mapping. These data substantially improve the accuracy and completeness of the reference sequence and the order and orientation of sequence scaffolds into chromosome arm assemblies. Representation of the Y chromosome and other heterochromatic regions is particularly improved. The new 143.9-Mb reference sequence, designated Release 6, effectively exhausts clone-based technologies for mapping and sequencing. Highly repeat-rich regions, including large satellite blocks and functional elements such as the ribosomal RNA genes and the centromeres, are largely inaccessible to current sequencing and assembly methods and remain poorly represented. Further significant improvements will require sequencing technologies that do not depend on molecular cloning and that produce very long reads.

Canu: scalable and accurate long-read assembly via adaptivek-mer weighting and repeat separation

“…Some chromosome arms are assembled into single contigs, but the graph reveals the structure of the more complex, unresolved repeats in the assembly. For example, Chromosome 2L is assembled as a single component in the graph but is broken toward the end due to a large array of transposable elements and the histone gene cluster, which spans >500 kbp (Hoskins et al 2015). These elements also correspond to unfinished gaps in the D. melanogaster reference.…”

Integrated tRNA, transcript, and protein profiles in response to steroid hormone signaling

“…Analysis of tRNA levels was restricted by the current annotation set for tRNA species using the reference 6 release of the Drosophila melanogaster genome (Hoskins et al 2015). To date, only 44 tRNA types, including the Selenocysteine isoacceptor, are currently annotated, whereas annotations for 19 tRNA types are currently missing, presenting a major limitation in our ability to estimate the abundance of all tRNA levels.…”

“…Small RNAsequencing reads and adaptor sequences were trimmed using the ngsShort toolkit (Chen et al 2014). Sequencing reads were subsequently mapped to the Dm6 release of the Drosophila genome (Hoskins et al 2015) using bowtie1.0.0 (Langmead 2010), allowing one mismatch and randomly assigning multimapping reads a single alignment to the highest scoring region. Exact read duplicates, which arise from PCR amplification were removed using samtools tRNA levels and protein translation www.rnajournal.org 1815 (Li et al 2009).…”

“…S1) and constructed indexed libraries for multiplex RNA sequencing (RNA-seq). RNA-seq reads were mapped to the release 6 reference sequence of the Drosophila melanogaster genome (Hoskins et al 2015), and the relative abundance for each tRNA and miRNA was determined by the enrichment of sequencing reads mapping to each annotated region.…”

Whole genome resequencing of a laboratory-adapted Drosophila melanogaster population sample

“…2057) 4, 5 , and have a different paternal genome each, sampled using cytogenetic cloning from the LH M base population. All non-reference genotypes in the sequenced LH M hemiclones were expected to be heterozygous and in-phase, except in rare instances where the in-house dm6 reference strain also had the same non-reference allele.…”