gmol.base.data.seq.mmseqs_search#

Some of the code is imported from ColabFold. See:

License

MIT License

Copyright (c) 2021 Sergey Ovchinnikov

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Classes

`MMseqs`([threads, skip_existing, mmseqs])
`MMseqsCommonSearchParams`(*[, ...])
`MMseqsMonomerSearchParams`(*[, ...])
`MMseqsPairSearchParams`(*[, prefilter_mode, ...])
`MMseqsQuery`(id, unique_seqs, seq_counts)

Functions

`mmseqs_search_monomer`(mmseqs, qdb, ...[, ...])	Run mmseqs with a local colabfold database set
`mmseqs_search_pair`(mmseqs, qdb, output_dir, ...)
`run_search_from_path`(query, output_dir[, ...])	Run MSA search from a FASTA/CSV/dir path.

class MMseqs(threads: int = 1, skip_existing: bool = True, mmseqs: str | Path = 'mmseqs')#

createdb(query_file: Path, qdb: Path, shuffle: int = 0, dbtype: int = 1)#

search(qdb: Path, db: Path, res: Path, tmp: Path, additional_search_param: Iterable[str | Path] | None = None)#

mvdb(src: Path, dest: Path)#

rmdb(db: Path)#

unpackdb(db: Path, dest: Path, unpack_name_mode: int = 0, unpack_suffix: str = '.a3m')#

result2msa(qdb: Path, db: Path, res: Path, dest: Path, msa_format_mode: int = 6, db_load_mode: int = 2, additional_params: Iterable[str | Path] | None = None)#

filterresult(qdb: Path, db: Path, res: Path, dest: Path, db_load_mode: int = 2, qid: str = '0', qsc: float = 0.8, diff: int = 0, max_seq_id: float = 1.0, filter_min_enable: int = 100)#

align(qdb: Path, db: Path, res: Path, dest: Path, db_load_mode: int = 2, align_eval: float | str = 10, max_accept: int = 1000000, alt_ali: int | None = 10, backtrace: bool = True)#

expandaln(qdb: Path, db: Path, res: Path, db2: Path, dest: Path, db_load_mode: int = 2, additional_expand_param: Iterable[str | Path] | None = None)#

lndb(src: Path, dest: Path)#

mergedbs(qdb: Path, dest: Path, *dbs: Path)#

convertalis(qdb: Path, db: Path, res: Path, dest: Path, db_load_mode: int = 2, format_output: str = 'query,target,fident,alnlen,mismatch,gapopen,qstart,qend,tstart,tend,evalue,bits,cigar', db_output: int = 1)#: Convert alignment result to tabular or DB. Skip if dest.dbtype exists.

pairaln(qdb: Path, db: Path, res: Path, dest: Path, db_load_mode: int = 2, pairing_mode: int = 0, pairing_dummy_mode: int = 0)#

class MMseqsQuery(id: str, unique_seqs: list[str], seq_counts: list[int])#

id: str#

unique_seqs: list[str]#

seq_counts: list[int]#

property heteromer: bool#

classmethod from_msa_query(msa_query: MsaQuery) → MMseqsQuery#

class MMseqsCommonSearchParams(*, prefilter_mode: int = 0, s: float | None = 8, gpu: int = 0, db_load_mode: int = 2, ignore_index: bool = True)#

prefilter_mode: int = 0#

s: float | None = 8#

gpu: int = 0#

db_load_mode: int = 2#

ignore_index: bool = True#

property search_param#

class MMseqsMonomerSearchParams(*, prefilter_mode: int = 0, s: float | None = 8, gpu: int = 0, db_load_mode: int = 2, ignore_index: bool = True, uniref_db: pathlib._local.Path, metagenomic_db: pathlib._local.Path | None, template_db: pathlib._local.Path | None, expand_eval: float = inf, align_eval: int = 10, diff: int = 3000, qsc: float = -20.0, max_accept: int = 1000000, template_s: float | None = 7.5, alt_ali: int = 10, limit: bool = True)#

Bases: MMseqsCommonSearchParams

uniref_db: Path#

metagenomic_db: Path | None#

template_db: Path | None#

expand_eval: float = inf#

align_eval: int = 10#

diff: int = 3000#

qsc: float = -20.0#

max_accept: int = 1000000#

template_s: float | None = 7.5#

alt_ali: int = 10#

limit: bool = True#

suffix1: str#

suffix2: str#

suffix3: str#

property template_search_param#

property filter_param#

property expand_param#

property udb1: Path#

property udb2: Path#

property mdb1: Path#

property mdb2: Path#

property tdb: Path#

class MMseqsPairSearchParams(*, prefilter_mode: int = 0, s: float | None = 8, gpu: int = 0, db_load_mode: int = 2, ignore_index: bool = True, search_db: pathlib._local.Path, pairing_strategy: int = 0)#

Bases: MMseqsCommonSearchParams

search_db: Path#

pairing_strategy: int = 0#

suffix1: str#

suffix2: str#

property expand_param#

property udb1: Path#

property udb2: Path#

mmseqs_search_monomer(mmseqs: MMseqs, qdb: Path, output_dir: Path, params: MMseqsMonomerSearchParams, temp_base: Path | None = None)#

Run mmseqs with a local colabfold database set

db1: uniprot db (UniRef30)
db3: metagenomic db (colabfold_envdb_202108 or bfd_mgy_colabfold, the former is preferred)

mmseqs_search_pair(mmseqs: MMseqs, qdb: Path, output_dir: Path, params: MMseqsPairSearchParams, unpack_suffix: str, temp_base: Path | None = None) → None#

run_search_from_path(query: Path, output_dir: Path, threads: int = 1, mmseqs: str | Path = 'mmseqs', temp_base: Path | None = None, *, monomer_params: MMseqsMonomerSearchParams, pair_params: MMseqsPairSearchParams, env_pair_params: MMseqsPairSearchParams | None = None) → None#

Run MSA search from a FASTA/CSV/dir path. Results written as output_dir/<jobname>.a3m and output_dir/<jobname>_<template_db>.m8 (if template search is enabled).

For complex (multi-chain) input, runs monomer unpaired + optional paired search and merges per job.