Source code for dammit.databases

# Copyright (C) 2015-2018 Camille Scott
# All rights reserved.
# This software may be modified and distributed under the terms
# of the BSD license.  See the LICENSE file for details.

import logging
import os
from os import path
import sys

from doit.dependency import Dependency, SqliteDB

from dammit import ui
from dammit.handler import TaskHandler
from dammit.tasks.last import LastDBTask
from dammit.tasks.hmmer import HMMPressTask
from dammit.tasks.infernal import CMPressTask
from import (get_download_and_gunzip_task,

[docs]def get_handler(config): '''Build the TaskHandler for the database prep pipeline. The handler will not have registered tasks when returned. Args: config (dict): Config dictionary, which contains the command line arguments and the entries from the config file. databases (dict): The database dictionary from `databases.json`. Returns: handler.TaskHandler: A constructed TaskHandler. ''' logger = logging.getLogger('DatabaseHandler') logger.debug('get_handler') handler = TaskHandler(config['database_dir'], logger, db='databases', backend=config['doit_backend'], verbosity=config['verbosity'], n_threads=config['n_threads']) return handler
[docs]def default_database_dir(logger): '''Get the default database directory: checks the environment for a DAMMIT_DB_DIR variable, and if it is not found, returns the default location of `$HOME/.dammit/databases`. Args: logger (logging.logger): Logger to write to. Returns: str: Path to the database directory. ''' try: directory = os.environ['DAMMIT_DB_DIR'] logger.debug('found DAMMIT_DB_DIR env variable') except KeyError: logger.debug('no DAMMIT_DB_DIR or --database-dir, using'\ 'default') directory = path.join(os.environ['HOME'], '.dammit', 'databases') return directory
[docs]def install(handler): '''Run the database prep pipeline from the given handler. ''' print(ui.header('Database Install', level=3)) print_meta(handler) msg = '*All database tasks up-to-date.*' uptodate, statuses = handler.print_statuses(uptodate_msg=msg) if not uptodate: print('Installing...') return else: print('Nothing to install!') return 0
[docs]def check_or_fail(handler): '''Check that the handler's tasks are complete, and if not, exit with status 2. ''' print(ui.header('Database Check', level=3)) print_meta(handler) msg = '*All database tasks up-to-date.*' uptodate, statuses = handler.print_statuses(uptodate_msg=msg) if not uptodate: print(ui.paragraph('Must install databases to continue. To do so,' ' run `dammit databases --install`. If you have' ' already installed them, make sure you\'ve given' ' the correct location to `--database-dir` or have' ' exported the $DAMMIT_DB_DIR environment' ' variable.')) sys.exit(2)
[docs]def build_default_pipeline(handler, config, databases, with_uniref=False): '''Register tasks for dammit's builtin database prep pipeline. Args: handler (handler.TaskHandler): The task handler to register on. config (dict): Config dictionary, which contains the command line arguments and the entries from the config file. databases (dict): The dictionary of files from `databases.json`. with_uniref (bool): If True, download and install the uniref90 database. Note that this will take 16+Gb of RAM and a looong time to prepare with `lastdb`. Returns: handler.TaskHandler: The handler passed in. ''' register_pfam_tasks(handler, config['hmmer']['hmmpress'], databases) register_rfam_tasks(handler, config['infernal']['cmpress'], databases) register_orthodb_tasks(handler, config['last']['lastdb'], databases) register_busco_tasks(handler, config, databases) if with_uniref: register_uniref90_tasks(handler, config['last']['lastdb'], databases) return handler
[docs]def build_quick_pipeline(handler, config, databases): register_busco_tasks(handler, config, databases) return handler
[docs]def register_pfam_tasks(handler, params, databases): pfam_A = databases['Pfam-A'] archive_fn = '{0}.{1}'.format(pfam_A['filename'], pfam_A['fileformat']) target_fn = path.join(, pfam_A['filename']) dl_task = get_download_task(pfam_A['url'], archive_fn, md5=pfam_A['md5']) gz_task = get_gunzip_task(archive_fn, target_fn) handler.register_task('download:Pfam-A', dl_task, files={'Pfam-A-gz': archive_fn}) handler.register_task('gunzip:Pfam-A', gz_task, files={'Pfam-A': target_fn}) handler.register_task('hmmpress:Pfam-A', HMMPressTask().task(target_fn, params=params)) return handler
[docs]def register_rfam_tasks(handler, params, databases): rfam = databases['Rfam'] archive_fn = '{0}.{1}'.format(rfam['filename'], rfam['fileformat']) target_fn = path.join(, rfam['filename']) dl_task = get_download_task(rfam['url'], archive_fn, md5=rfam['md5']) gz_task = get_gunzip_task(archive_fn, target_fn) handler.register_task('download:Rfam', dl_task, files={'Rfam-gz': archive_fn}) handler.register_task('gunzip:Rfam', gz_task, files={'Rfam': target_fn}) handler.register_task('cmpress:Rfam', CMPressTask().task(target_fn, params=params)) return handler
[docs]def register_orthodb_tasks(handler, params, databases): orthodb = databases['OrthoDB'] archive_fn = '{0}.{1}'.format(orthodb['filename'], orthodb['fileformat']) target_fn = path.join(, orthodb['filename']) dl_task = get_download_task(orthodb['url'], archive_fn, md5=orthodb['md5']) gz_task = get_gunzip_task(archive_fn, target_fn) handler.register_task('download:OrthoDB', dl_task, files={'OrthoDB-gz': archive_fn}) handler.register_task('gunzip:OrthoDB', gz_task, files={'OrthoDB': target_fn}) handler.register_task('lastdb:OrthoDB', LastDBTask().task(target_fn, target_fn, prot=True, params=params)) return handler
[docs]def register_busco_tasks(handler, config, databases): busco = databases['BUSCO'] busco_dir = path.join(, config['busco']['db_dir']) group_name = config['busco_group'] group = busco[group_name] files = {'BUSCO-{0}'.format(group_name): path.join(busco_dir, group['folder'])} handler.register_task('download:BUSCO-{0}'.format(group_name), get_download_and_untar_task(group['url'], busco_dir, label=group_name), files=files) return handler
[docs]def register_uniref90_tasks(handler, params, databases): uniref90 = databases['uniref90'] task = get_download_and_gunzip_task(uniref90['url'], uniref90['filename']) filename = path.join(, uniref90['filename']) handler.register_task('download:uniref90', task, files={'uniref90': filename}) handler.register_task('lastdb:uniref90', LastDBTask().task(filename, filename, prot=True, params=params, task_dep=[])) return handler