search

abutils provides functions for searching sequences against a database of target sequences using MMseqs2. This allows for fast and efficient similarity search, which is useful for tasks like sequence identification, annotation, and homology detection.

search method	function
MMseqs2	abutils.tools.mmseqs_search()

examples

search with MMseqs2

The MMseqs2 search function can accept a path to a FASTA file, an MMseqs2 database, a abutils.Sequence object, or an iterable of abutils.Sequence objects for both query and target sequences.

import abutils

# search sequences against a target database
results = abutils.tools.mmseqs_search(
    query='path/to/query_sequences.fasta',
    target='path/to/target_sequences.fasta',
    output_path='path/to/output.tsv'
)

customize search parameters

MMseqs2 search can be customized with various parameters to control sensitivity, format, and performance.

import abutils

# search with customized parameters
results = abutils.tools.mmseqs_search(
    query='path/to/query_sequences.fasta',
    target='path/to/target_sequences.fasta',
    output_path='path/to/output.tsv',
    search_type=1,  # amino acid search
    max_seqs=100,   # maximum hits per query
    max_evalue=1e-5,  # stricter E-value cutoff
    sensitivity=7.5,  # higher sensitivity
    format_mode=4,    # BLAST-TAB + column headers
    threads=8         # use 8 threads
)

customizing output format

You can customize the output format to include specific columns.

import abutils

# customize the output format
results = abutils.tools.mmseqs_search(
    query='path/to/query_sequences.fasta',
    target='path/to/target_sequences.fasta',
    output_path='path/to/output.tsv',
    format_mode=4,  # BLAST-TAB + column headers
    format_output="query,target,evalue,pident,qcov,tcov"
)

api