Profile hidden Markov model sequence analysis can help remove putative pseudogenes from DNA barcoding and metabarcoding datasets

Abstract

BackgroundPseudogenes are non-functional copies of protein coding genes that typically follow a different molecular evolutionary path as compared to functional genes. The inclusion of pseudogene sequences in DNA barcoding and metabarcoding analysis can lead to misleading results. None of the most widely used bioinformatic pipelines used to process marker gene (metabarcode) high throughput sequencing data specifically accounts for the presence of pseudogenes in protein-coding marker genes. The purpose of this study is to develop a method to screen for obvious pseudogenes in large COI metabarcode datasets. We do this by: 1) describing gene and pseudogene characteristics from a simulated DNA barcode dataset, 2) show the impact of two different pseudogene removal methods on mock metabarcode datasets with simulated pseudogenes, and 3) incorporate a pseudogene filtering step in a bioinformatic pipeline that can be used to process Illumina paired-end COI metabarcode sequences. Open reading frame length and sequence bit scores from hidden Markov model (HMM) profile were used to detect pseudogenes.ResultsOur simulations showed that it was more difficult to identify pseudogenes from shorter amplicon sequences such as those typically used in metabarcoding (∼300 bp) compared with full length DNA barcodes that are used in construction of barcode libraries (∼ 650 bp). It was also more difficult to identify pseudogenes in datasets where there is a high percentage of pseudogene sequences. We show that existing bioinformatic pipelines used to process metabarcode sequences already remove some apparent pseudogenes, especially in the rare sequence removal step, but the addition of a pseudogene filtering step can remove more.ConclusionsThe combination of open reading frame length and hidden Markov model profile analysis can be used to effectively screen out obvious pseudogenes from large datasets. There is more to learn from COI pseudogenes such as their frequency in DNA barcode and metabarcoding studies, their taxonomic distribution, and evolution. Thus, we encourage the submission of verified COI pseudogenes to public databases to facilitate future studies.

Do pseudogenes pose a problem for metabarcoding marine animal communities?

“…Our results indicate that the impacts of NUMTs can be greatly reduced by targeting longer amplicons because most NUMTs are short. Moreover, NUMTs <300 bp are less likely to contain diagnostic features such as inappropriate amino acid substitutions that can be identified via bioinformatic pipelines (Porter & Hajibabaei, 2021). Accordingly, NUMTs pose the highest risk to eDNA studies where amplicons typically range from 50 to 200 bp (Langlois et al, 2021) to dietary analyses where amplicons are 70–230 bp (Berry et al, 2017; da Silva et al, 2019), and to processed fish samples which commonly use amplicons of 127–314 bp (Shokralla et al, 2015).…”

“…The impact of unrecognized NUMTs on diversity estimates will differ among such approaches. Whether OTUs are or are not used, sequence arrays containing NUMTs can still accurately represent beta diversity if NUMTs are either uncommon or consistently recovered across samples (Porter & Hajibabaei, 2021).…”

Towards eradicating the nuisance of numts and noise in molecular biodiversity assessment

“…Nevertheless it is a key utility that can easily be included in metabarcode data processing pipelines and importantly provides detailed results to evaluate data set characteristics. We were pleased to recently see another contribution to the literature arriving for peer‐review that uses a combination of open reading frame length and hidden Markov model profile analysis for numt removal (Porter & Hajibabaei, 2021). Those authors encourage the submission of verified cytochrome oxidase subunit I (COI) pseudogenes to public databases to facilitate future studies.…”

Multi-marker metabarcoding resolves subtle variations in freshwater condition: Bioindicators, ecological traits, and trophic interactions

“…As we were using protein coding markers in this study, we also screened out obvious pseudogenes to try to reduce noise in the dataset and avoid inflating richness estimates 103 . For rbcL, we removed putative pseudogenes using removal method 1:…”