README.md

In fishsciences/metabarcode_utils: Meta-barcoding Utilities

metabarcodeUtils

This simple package contains helper functions for the meta-barcoding workflow.

Installing the package

devtools::install_github("fishsciences/metabarcode_utils")

Using the package

Summarizing MitoFish Output

This package currently starts with a xlsx result file, renamed, from the MitoFish tool. For testing, an example file is included with the package.

f = system.file(package = "metabarcodeUtils",
                "sample_files",
                "BTWIT1-20200429B.xlsx")

This file can be parsed and converted into two CSV files (one for the hits and one for the no hits) using the mitofish_to_csv() function:

tt = mitofish_to_csv(f, outdir = out_dir)

By default, this will return a list with two data.frames, and will write the results to outdir. These results can then be further processed into Fasta files with csv_to_fasta():

csv_to_fasta(file.path(out_dir, ff[1]), outdir = out_dir)

Summarizing NCBI BLAST Output

To extract and summarize the tables from a NCBI output file,

ncbi_tabs = extract_ncbi_tabs("R69MU8R7016-Alignment.txt")

This returns all the values in the table. To subset these according to:

• Step 1. Query cover matters. If coverage is less than 90% (this may be too high, can we make this adjustable?) then we cannot identify the sequence to any organism. The output should be “<90% coverage” for this “Query”

• Step 2. Is Per Ident >97%. If yes, then we need Scientific name, common name, query cover, Per ID, Accession#. We only need one record of each species.

• Step 3. Is Per Ident >97% for multiple species. If yes, then we need Scientific name, common name, query cover, Per ID, Accession#. We only need one record of each species.

• Step 4. Is Per Ident between 80-96.9%. If yes, then we need Scientific name, common name, query cover, Per ID, Accession# for all records. We only need one record for each species.

• Step 5. Is Per Ident less than 80%. If yes, then output should be “<80% per ident”

This is accomplished with

top_accessions = reduce_ncbi(ncbi_tabs, accession_keep = "top")

These tables can be written to a CSV file with,

d = create.dir("ncbi_output")
write_ncbi_tabs(top_accessions, "ncbi_output")

fishsciences/metabarcode_utils documentation built on Feb. 25, 2024, 11:31 a.m.