TABLE 1.
Table of all bioinformatic tasks performed across the core papers set
| Task group | Task | Description | Number of papers reporting task | Number of papers not reporting software | Total number of software tools | Total number of software functions | Number of papers performing manually |
|---|---|---|---|---|---|---|---|
| Read preparation | Quality control | Generating a report of sequence quality information from a sample or set of samples ‐ no modification is done to data | 19 | 0 | 4 | 4 | 0 |
| Adapter trimming | Trimming of sequencing adapters | 9 | 1 | 6 | 6 | 0 | |
| Demultiplexing | Separation of sequences from a mixed pool into separate pools based on the occurrence of a unique set of bases (index or tag) | 55 | 17 | 16 | 19 | 0 | |
| Pair merging | The assembly of mate pair reads into a single contig | 63 | 1 | 10 | 18 | 0 | |
| Quality trimming | The removal of bases from either or both ends of sequences in a pool based on quality scores | 20 | 1 | 8 | 10 | 0 | |
| Mate pairing | The identification and syncronisation of mate pair reads between two samples, often involving arranging reads in identical orders and/or removal of reads without a mate pair | 3 | 0 | 3 | 3 | 0 | |
| Primer trimming | Trimming of PCR primers | 66 | 8 | 15 | 17 | 0 | |
| Reverse complementation | Reverse complementing the sequences in a pool | 7 | 3 | 2 | 2 | 0 | |
| Sequence conversion | Converting sequences from fastq to fasta | 3 | 0 | 2 | 3 | 0 | |
| Length trimming | The removal of bases from either or both ends of sequences in a pool, either the removal of a fixed number of bases or the removal of a variable number of bases to reduce sequences to a standard length | 10 | 3 | 6 | 7 | 0 | |
| Pair concatenation | Concatenating mate pair reads into a single contig (where reads don't overlap) | 8 | 4 | 4 | 4 | 0 | |
| Assembly | The assembly of reads into contigs, applied when more than one pair of overlapping fragments have been metabarcoded | 6 | 0 | 4 | 4 | 0 | |
| Degapping | Removal of gaps from sequences | 1 | 0 | 1 | 1 | 0 | |
| Sequence processing | Dereplication | The removal of duplicate reads to retain only unique sequences in a pool; often the total number of copies of a sequence is recorded in the header of the retained sequence | 58 | 10 | 11 | 19 | 0 |
| Size sorting | The sorting of a fasta file according to a size annotation in the header | 10 | 2 | 3 | 4 | 0 | |
| Filtering | Quality filtering | Removal and/or trimming of sequences from a pool based on quality information. Also often converts from fastq to fasta. | 81 | 11 | 20 | 27 | 0 |
| Similarity filtering | Removal of sequences based on similarity to an alignment, either based on sequence identity or alignment position | 9 | 1 | 4 | 4 | 0 | |
| Length filtering | The removal of sequences from a pool that are less than, more than, or fall within or outside of a specified length threshold or thresholds | 54 | 21 | 17 | 23 | 0 | |
| Preclustering | Reduction of sequence variation in a dataset prior to further processing ‐ a form of denoising | 12 | 1 | 3 | 6 | 0 | |
| Denoising | The removal of reads containing putative PCR or sequencing errors based on statistical assessment | 18 | 1 | 8 | 8 | 0 | |
| Normalisation | A process by which the number of sequences for each of a set of samples is reduced where necessary such that the output set of samples all have the same number of sequences while maintaining the relative frequencies of OTUs | 2 | 0 | 1 | 1 | 1 | |
| Chimera filtering | The filtering of putative chimeric assemblies from a pool of mate paired reads | 63 | 4 | 6 | 16 | 1 | |
| Translation filtering | Removal of sequences from a set of sequence based on their translation, usually removing sequences with inframe stop codons or frameshifts due to erroneous indels or substitutions caused by sequencing errors | 22 | 3 | 11 | 12 | 0 | |
| Frequency filtering | Removal of sequences based on their frequency in a pool | 51 | 37 | 11 | 15 | 1 | |
| Taxonomy filtering | Removal of sequences based on an assigned taxonomy or a taxonomic classification | 9 | 5 | 1 | 1 | 1 | |
| Mistag filtering | Removal of sequences based on putative tagging errors | 3 | 1 | 1 | 1 | 0 | |
| Data generation | OTU delimitation | The grouping of a set of sequences into OTUs by some method | 84 | 5 | 12 | 22 | 0 |
| OTU mapping | The mapping of sequences to OTUs to provide read counts for each OTU | 30 | 3 | 7 | 11 | 0 | |
| Uncurated taxonomic assignment | The assignment (identification or classification) of taxonomy to OTUs using a global uncorated reference database (e.g., GenBank, BOLD) | 55 | 2 | 11 | 13 | 0 | |
| Reference taxonomic assignment | The assignment (identification or classification) of taxonomy to OTUs using a purpose‐built and/or specially curated reference set of sequences | 60 | 9 | 18 | 23 | 1 |
Tasks are grouped into four groups by broad purpose, and a detailed definition of each task is given along with summary statistics of the implementation of each task across the 111 papers. For a list of the software used for each task, Table S1 is an expanded version of this table.