Skip to content

Nucleic Search

Summary

Search nucleotide reads or contigs against virus reference databases.

Description

Input can be one FASTA/FASTQ file, a comma-separated list, a directory, or an existing MMseqs2 database. Sequence inputs are combined into one MMseqs2 query database. Records are searched independently; paired-read concordance is not evaluated here. Use rolypoly map for pair-aware read mapping.

For a custom nucleotide reference distributed as FASTA (including gzip-compressed FASTA), use --db other --db-path reference.fasta.gz; RolyPoly will create the temporary MMseqs2 target database. This can also serve as a workaround when an older installation points to an obsolete built-in database path.

Usage

rolypoly nucleic-search [OPTIONS]

Options

  • -o, --output: output file location - set suffix to .tab, .sam or html (type: TEXT; default: /home/neri/Documents/Github/rolypoly_RP_mapping)
  • --db, --database: Select the database to search against. 'all' retains its historical meaning: the two RNA-virus databases (RVMT and NCBI_Ribovirus). (type: CHOICE; default: all)
  • --db-path: Path to the user-supplied source (required if --db is 'other'). Either a fasta or a path to formatted MMseqs2 virus database (type: TEXT; default: ``)
  • -i, --input: Input FASTA/FASTQ file, comma-separated sequence files, directory of sequence files, or one preformatted MMseqs2 database prefix (type: TEXT; required; default: Sentinel.UNSET)
  • -mo, --matched-output: Output path for matched virus contigs. set to 'no' to skip writing matched contigs (type: TEXT; default: Sentinel.UNSET)
  • -e, --mmseqs-evalue: E-value threshold for MMseqs2 search) (type: FLOAT; default: 0.1)
  • -id, --mmseqs-identity: minimum Identity threshold for MMseqs2 search) (type: FLOAT; default: 0.7)
  • -al, --mmseqs-min-aln-len: Minimum alignment length for MMseqs2 search) (type: INTEGER; default: 95)
  • -t, --threads: Number of worker threads. (type: INTEGER RANGE; default: 1)
  • -M, --memory: Memory limit, for example 8g. (type: MEMORY; default: 8g)
  • -k, --keep-tmp: Keep temporary files. (type: BOOLEAN; default: False)
  • -tmp, --temp-dir: Temporary working directory. (type: DIRECTORY)
  • -g, --log-file: Path to the log file. (type: FILE; default: rolypoly.log)

Caveats

Nucleic hits may end at about 10 kbp because MMseqs2 internally splits long sequences in its nucleotide-search workflow. This is not evidence of a biological boundary. See the report caveat for interpretation and upstream implementation context.

Known bugs

The report does not reconstruct continuous alignments across internal sequence splits. See report known bugs.