Terminating contamination: large-scale search identifies more than 2,000,000 contaminated entries in GenBank

Abstract

<jats:p>Metagenomic sequencing allows researchers to investigate organisms sampled from their native environments by sequencing their DNA directly, and then quantifying the abundance and taxonomic composition of the organisms thus captured. However, these types of analyses are sensitive to contamination in public databases caused by incorrectly labeled reference sequences. Here we describe Conterminator, an efficient method to detect and remove incorrectly labelled sequences by an exhaustive all-against-all sequence comparison. Our analysis reports contamination in 114,035 sequences and 2767 species in the NCBI Reference Sequence Database (RefSeq), 2,161,746 sequences and 6795 species in the GenBank database, and 14,132 protein sequences in the NR non-redundant protein database. Conterminator uncovers contamination in sequences spanning the whole range from draft genomes to “complete” model organism genomes. Our method, which scales linearly with input size, was able to process 3.3 terabytes of genomic sequence data in 12 days on a single 32-core compute node. We believe that Conterminator can become an important tool to ensure the quality of reference databases with particular importance for downstream metagenomic analyses. Source code (GPLv3): <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="https://github.com/martin-steinegger/conterminator">https://github.com/martin-steinegger/conterminator</jats:ext-link></jats:p>

Accurate and sensitive detection of microbial eukaryotes from whole metagenome shotgun sequencing

“…Eukaryotic reference genomes and proteomes have many sequences that are derived from bacteria, which have entered these genomes either spuriously through contamination during sequencing and assembly, or represent true biology of horizontal transfer from bacteria to eukaryotes [22,23]. In either case, these bacterial-derived sequences overwhelm the ability of either k -mer matching or read mapping-based approaches to detect eukaryotes from microbiome sequencing, as bacteria represent the majority of the sequencing library from many microbiomes.…”

“…However, databases of eukaryotic genomes and proteins have widespread contamination from bacterial sequences, and these methods therefore frequently misattribute bacterial reads to eukaryotic species [22,23]. Gene-based taxonomic profilers, such as Metaphlan3, have been developed to detect eukaryotic species, but these target a small subset of microbial eukaryotes (122 eukaryotic species as of the mpa_v30 release) [24,25].…”

“…Research questions about eukaryotes in microbiomes using whole metagenome sequencing have primarily been addressed by using eukaryotic reference genomes, genes, or proteins for either k -mer matching or read mapping approaches [20,21]. However, databases of eukaryotic genomes and proteins have widespread contamination from bacterial sequences, and these methods therefore frequently misattribute bacterial reads to eukaryotic species [22,23]. Gene-based taxonomic profilers, such as Metaphlan3, have been developed to detect eukaryotic species, but these target a small subset of microbial eukaryotes (122 eukaryotic species as of the mpa_v30 release) [24,25].…”

Terminating contamination: large-scale search identifies more than 2,000,000 contaminated entries in GenBank

“…Conterminator [35] is implemented in C++ and its open source licensed as GPLv3 and available at https://github.com/ martin-steinegger/conterminator. The version to reproduce the results is available under https://doi.org/10.5281/ zenodo.3750825 [36]. Commands to rerun the analysis of RefSeq and NR are in Additional file 3: Listing S1.…”

SkewIT: Skew Index Test for detecting mis-assembled bacterial genomes

“…For sequences to be entered into RefSeq, curators at NCBI perform both automated and manual checks to ensure minimal contamination and high sequence quality. Despite these efforts, multiple studies have identified contamination in RefSeq and other publicly available genome databases [3,4,5,6,7]. NCBI requires Refseq assemblies to have an appropriate genome length as compared to existing genomes from the same species, and it labels assemblies as "complete" if the genome exists in one contiguous sequence per chromosome, with no unplaced scaffolds and with all chromosomes present.…”