DENTIST – using long reads for closing assembly gaps at high accuracy

Abstract

Long sequencing reads allow increasing contiguity and completeness of fragmented, short-read based genome assemblies by closing assembly gaps, ideally at high accuracy. While several gap closing methods have been developed, these methods often close an assembly gap with sequence that does not accurately represent the true sequence. Here, we developed DENTIST, a sensitive, highly-accurate and automated pipeline method to close gaps in short read assemblies with long reads. DENTIST comprehensively determines repetitive assembly regions to identify reliable and unambiguous alignments of long reads to the right loci, integrates a consensus sequence computation step to obtain a high base accuracy for the inserted sequence, and validates the accuracy of closed gaps. Unlike previous benchmarks, we generated test assemblies that have gaps at the exact positions where real short-read assemblies have gaps. Generating such realistic benchmarks for Drosophila (134 Mb genome), Arabidopsis (119 Mb), hummingbird (1 Gb) and human (3 Gb) and using simulated or real PacBio reads, we show that DENTIST consistently achieves a substantially higher accuracy compared to previous methods, while having a similar sensitivity. As another distinguishing feature, DENTIST can accurately scaffold the given contigs with long reads in addition to closing gaps, extending its application range to contig-only assemblies. In summary, DENTIST provides an accurate approach to improve the contiguity and completeness of fragmented assemblies with long reads. DENTIST's source code including a Snakemake workflow and Docker container is available at https://github.com/a-ludi/dentist. All test assemblies as a resource for future benchmarking are at https://bds.mpi-cbg.de/hillerlab/DENTIST/.

DENTIST—using long reads for closing assembly gaps at high accuracy

“…Furthermore, we distributed the computation on our compute cluster by (i) computing the read alignments for the whole assembly, (ii) splitting the assembly into blocks of ∼200 Mb and dividing the read alignments accordingly, (iii) applying PacBio GC to each block separately, and (iv) merging the unmodified contigs with the processed gaps into the output assembly. This complete workflow can be found at [ 31, 32 ].…”

“…All data underlying this article, including the reference and test assemblies with introduced gaps and their true sequence as valuable data for future method comparisons, are available via our institutional server [ 31 ]. Supporting data and an archival copy of the code are also available via the GigaScience repository GigaDB [ 33 ].…”

A Deep Dive into Genome Assemblies of Non-vertebrate Animals

“…LoRMA [150], NaS [151], proovread [152] Long reads Canu [114], CONSENT [153], Daccord [154], FLAS [155], HALC [156], NextDenovo [121], MECAT [118], MECAT2 [118], NECAT [120] Polishing Short reads ntEdit [157], Pilon [158], POLCA [159] Short & long reads Apollo [160], Hapo-G [161], HyPo [162], Racon [163] Nanopolish [164], Quiver [165] Haplotig purging Long reads HaploMerger2 [166], purge dups [167], Purge Haplotigs [168] Scaffolding Short reads Bambus [169], BATISCAF [170], BESST [171], BOSS [172], Mate pairs GRASS [173], MIP [174], Opera [175], ScaffMatch [176], ScaffoldScaffolder [177], SCARPA [178], SCOP [179], SLIQ [180], SOPRA [181], SSPACE [182], WiseScaffolder [183] Long reads DENTIST [184], LINKS [185], LRScaf [186], npScarf [187], PBJelly [188], RAILS [189], SLR [190],...…”

“…Cobbler [189], DENTIST [184], FGAP [219], GMcloser [220], LR Gapcloser [221], PBJelly [188], PGcloser [222], TGS-GapCloser [223] thoroughly reviewed on multiple datasets [225]. When tested on Caenorhabditis elegans Nanopore reads, the error rate decreased from 28.93% to less than 1% (using Canu [114], CONSENT [153], FLAS [155], Jabba [148],…”