Alignment-free classification of COI DNA barcode data with the Python package Alfie

Abstract

Characterization of biodiversity from environmental DNA samples and bulk metabarcoding data is hampered by off-target sequences that can confound conclusions about a taxonomic group of interest. Existing methods for isolation of target sequences rely on alignment to existing reference barcodes, but this can bias results against novel genetic variants. Effectively parsing targeted DNA barcode data from off-target noise improves the quality of biodiversity estimates and biological conclusions by limiting subsequent analyses to a relevant subset of available data. Here, we present Alfie, a Python package for the alignment-free classification of cytochrome c oxidase subunit I (COI) DNA barcode sequences to taxonomic kingdoms. The package determines k-mer frequencies of DNA sequences, and the frequencies serve as input for a neural network classifier that was trained and tested using ~58,000 publicly available COI sequences. The classifier was designed and optimized through a series of tests that allowed for the optimal set of DNA k-mer features and optimal machine learning algorithm to be selected. The neural network classifier rapidly assigns COI sequences of varying lengths to kingdoms with greater than 99% accuracy and is shown to generalize effectively and make accurate predictions about data from previously unseen taxonomic classes. The package contains an application programming interface that allows the Alfie package’s functionality to be extended to different DNA sequence classification tasks to suit a user’s need, including classification of different genes and barcodes, and classification to different taxonomic levels. Alfie is free and publicly available through GitHub (https://github.com/CNuge/alfie) and the Python package index (https://pypi.org/project/alfie/).

Comparing the performance of 12S mitochondrial primers for fish environmental DNA across ecosystems

“…Neural networks have been shown to perform well for taxonomic assignments of microbial sequences based on patterns within the DNA sequence (Busia et al, 2019) and can help test taxonomic assignment abilities at various taxonomic levels. Generally, the flexibility of neural networks and their ability to learn complex patterns from large numbers of short sequences make them a promising choice for solving the general sequence‐labeling problem in eDNA research (Nugent & Adamowicz, 2020).…”

Molecular Acquisition, Cleaning and Evaluation in R (MACER) - A tool to assemble molecular marker datasets from BOLD and GenBank

“…Similarly, an alternative align_to_ref() could be written, using the current function as a model, utilising a different alignment algorithm, while keeping it incorporated into the current workflow. Finally, while the barcode_clean() function does currently assess the data, based on a number of metrics, including statistical outliers and amino acid translation, there are additional methods for assessing molecular sequence data and flagging potentially inaccurate data points ( Zhang et al 2012 , Nugent and Adamowicz 2020 , Fontes et al 2021 ). With some of these being built in R, the integration of these additional cleaning and verification methods can be placed into the MACER pipeline.…”

Efficacy and accuracy responses of DNA mini-barcodes in species identification under a supervised machine learning approach

“…CAOS [40], BLOG [10], DNABar [41], BRONX [27], DOME-ID [27] are some instances of this approach. There are alignment-free tools such as: ATIM-TNT (Tree-based) [27], CVTreeAlpha1.0 (Component Vector based) [42], Spectrum Kernel [43] and Alfie (python based) [44]. Web Based Tools such as: Linker [45], iBarcode [46], BioBarcode [47] and ConFind [48] are also there for DNA barcoding.…”