MEGAN-LR: New algorithms allow accurate binning and easy interactive exploration of metagenomic long reads and contigs

Abstract

Background There are numerous computational tools for taxonomic or functional analysis of microbiome samples, optimized to run on hundreds of millions of short, high quality sequencing reads.Programs such as MEGAN allow the user to interactively navigate these large datasets. Long read sequencing technologies continue to improve and produce increasing numbers of longer reads (of varying lengths in the range of 10k-1M bps, say), but of low quality. There is an increasing interest in using long reads in microbiome sequencing and there is a need to adapt short read tools to long read datasets. MethodsWe describe a new LCA-based algorithm for taxonomic binning, and an interval-tree based algorithm for functional binning, that are explicitly designed for long reads and assembled contigs. We provide a new interactive tool for investigating the alignment of long reads against reference sequences.For taxonomic and functional binning, we propose to use LAST to compare long reads against the NCBInr protein reference database so as to obtain frame-shift aware alignments, and then to process the results using our new methods.Results All presented methods are implemented in the open source edition of MEGAN and we refer to this new extension as MEGAN-LR (MEGAN long read). We evaluate the LAST+MEGAN-LR approach * To whom correspondence should be addressed.1 in a simulation study, and on a number of mock community datasets consisting of Nanopore reads, PacBio reads and assembled PacBio reads. We also illustrate the practical application on a Nanopore dataset that we sequenced from an anammox bio-rector community. BackgroundThere are numerous computational tools for taxonomic or functional binning or profiling of microbiome samples, optimized to run on hundreds of millions of short, high quality sequencing reads [1,2,3,4].Alignment-based taxonomic binning is often performed using the naïve LCA algorithm [5], because it is fast, easy to interpret and easy to implement. Functional binning usually involves a best-hit strategy to assign reads to functional classes.Software or websites for analyzing microbiome shotgun sequencing samples usually provide some level of interactivity, such as MG-RAST [2]. The interactive microbiome analysis tool MEGAN, which was first used in 2006 [6], is one of the most feature-rich tools of this type. MEGAN is highly optimized to enable users to interactively explore large numbers of microbiome samples containing hundreds of millions of short reads.Illumina HiSeq and MiSeq sequencers allow researchers to generate sequencing data on a huge scale, so as to analyze many samples at a great sequencing depth [7,8,9]. A wide range of questions, in particular involving the presence or absence of particular organisms or genes in a sample, can be answered using such data. However, there are interesting problems that are not easily resolved using short reads. For example, the question whether two genes, which both are detected in the same microbiome sample, also occur together on the same genome, can often not be ...

Optimizing extraction and targeted capture of ancient environmental DNA for reconstructing past environments using the PalaeoChip Arctic-1.0 bait-set

“…For example, to call Bison bison as being present in this instance, at least 3 unique reads must align to a region of the B. bison mitogenome with ≥ 95% identical similarity, with low e-values and high bit scores (metrics used by BLASTn to assess the likelihood of misalignments), and by matching or exceeding other parameters used to define confidence in the taxon identification (the rational for select LCA parameters are discussed further in supplementary Appendix A SET-E. Bioinformatic workflow). Adjusting LCA parameters shifts the trade-off ratio of false positive to false negative assignments, although it would seem that optimal LCA parameters may only exist on a project-by-project or even sample-bysample basis depending on the taxonomic molecular constituents present, the degree of aDNA damage, and the research question (see Huson et al, 2018). For example, if percent identity is set to 100, only exact matches will be considered, but then aDNA fragments with terminal base modifications (the majority of aDNA molecules) will be unassigned when their taxonomic classification might otherwise be obvious.…”

Long-Read Metagenomics to Retrieve High-Quality Metagenome-Assembled Genomes from Canine Feces

“…The next step for the HQ MAGs was to correct the frameshift errors, as described in [15], using Diamond 0.9.32 [27] and MEGAN-LR 6.19.1 [28]. We used ideel [29] to visualize the number of truncated ORF.…”

Gut microbiome of endangered Tor putitora (Ham.) as a reservoir of antibiotic resistance genes and pathogens associated with fish health

“…For functional characterization, raw reads were submitted for BLAST search against the nr database using diamond BLASTx at e-value 1e-3, similarity > 90%, and alignment length > 20 amino acids [111]. Using the best-hit algorithm, individual reads were described to belong to a class in the particular classification system [112]. The method was as follows: For a read 'r', let 'a' describe the highest-scoring alignment to a reference protein belonging to functional class 'c' and the number of reads that mapped to the individual proteins were then analysed using databases eggNOG [113], KEGG [114] and SEED [115].…”