query.py

# -*- coding: utf-8 -*-

from .database import BaseDbManager
from . import models
from sqlalchemy import distinct
from pandas import read_sql
from collections import Iterable


class QueryManager(BaseDbManager):
    """Query interface to database."""

    def _limit_and_df(self, query, limit, as_df=False):
        """adds a limit (limit==None := no limit) to any query and allow a return as pandas.DataFrame

        :param bool as_df: if is set to True results return as pandas.DataFrame
        :param `sqlalchemy.orm.query.Query` query: SQL Alchemy query 
        :param int,tuple limit: maximum number of results
        :return: query result of pyctd.manager.models.XY objects
        """
        if limit:

            if isinstance(limit, int):
                query = query.limit(limit)

            if isinstance(limit, Iterable) and len(limit) == 2 and [int, int] == [type(x) for x in limit]:
                page, page_size = limit
                query = query.limit(page_size)
                query = query.offset(page * page_size)

        if as_df:
            results = read_sql(query.statement, self.engine)

        else:
            results = query.all()

        return results

    @classmethod
    def _model_query(cls, query_obj, search4, model_attrib):

        if isinstance(search4, str):
            query_obj = query_obj.filter(model_attrib.like(search4))
        elif isinstance(search4, int):
            query_obj = query_obj.filter(model_attrib == search4)
        elif isinstance(search4, Iterable):
            query_obj = query_obj.filter(model_attrib.in_(search4))
        return query_obj

    def get_obo_string(self, taxid=None, limit=None):

        q = self.session.query(models.Entry)

        if limit:
            q = q.limit(limit)
        if taxid:
            q = q.filter(models.Entry.taxid == taxid)

        obo_string = ''

        for entry in q.all():
            obo_string += '\n[Term]\nid: SWISSPROT:{}\n'.format(entry.accessions[0])

            if len(entry.accessions) > 1:
                for accession in entry.accessions[1:]:
                    obo_string += 'alt_id: {}\n'.format(accession)

            obo_string += 'name: {}\n'.format(entry.recommended_full_name)

            for alternative_name in entry.alternative_full_names     + entry.alternative_short_names:
                obo_string += 'synonym: "{}" EXACT ALTERNATIVE_NAME []\n'.format(alternative_name.name)

            obo_string += 'synonym: "{}" EXACT GENE_NAME []\n'.format(entry.gene_name)

            for xref in entry.db_references:
                if xref.type_ in ['GO', 'HGNC']:
                    xref.identifier = ':'.join(xref.identifier.split(':')[1:])
                    obo_string += 'xref: {}:{}\n'.format(xref.type_, xref.identifier.replace('\\', '\\\\'))
        return obo_string

    def get_model_queries(self, query_obj, model_queries_config):
        for search4, model_attrib in model_queries_config:

            if search4 is not None:
                query_obj = self._model_query(query_obj, search4, model_attrib)
        return query_obj

    def get_one_to_many_queries(self, query_obj, one_to_many_queries):
        for search4, model_attrib in one_to_many_queries:
            if search4 is not None:
                query_obj = self._one_to_many_query(query_obj, search4, model_attrib)
        return query_obj

    def get_many_to_many_queries(self, query_obj, many_to_many_queries_config):
        for search4, model_attrib, many2many_attrib in many_to_many_queries_config:
            if search4 is not None:
                query_obj = self._many_to_many_query(query_obj, search4, model_attrib, many2many_attrib)
        return query_obj

    @classmethod
    def _many_to_many_query(cls, query_obj, search4, join_attrib, many2many_attrib):

        if isinstance(search4, str):
            query_obj = query_obj.join(join_attrib).filter(many2many_attrib.like(search4))

        elif isinstance(search4, int):
            query_obj = query_obj.join(join_attrib).filter(many2many_attrib == search4)

        elif isinstance(search4, Iterable):
            query_obj = query_obj.join(join_attrib).filter(many2many_attrib.in_(search4))

        return query_obj

    @classmethod
    def _one_to_many_query(cls, query_obj, search4, model_attrib):
        """extends and returns a SQLAlchemy query object to allow one-to-many queries

        :param query_obj: SQL Alchemy query object
        :param str search4: search string
        :param model_attrib: attribute in model
        """
        model = model_attrib.parent.class_

        if isinstance(search4, str):
            query_obj = query_obj.join(model).filter(model_attrib.like(search4))

        elif isinstance(search4, int):
            query_obj = query_obj.join(model).filter(model_attrib == search4)

        elif isinstance(search4, Iterable):
            query_obj = query_obj.join(model).filter(model_attrib.in_(search4))

        return query_obj

    def keyword(self, name=None, identifier=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.Pmid`

        .. seealso::

            :class:`pyuniprot.manager.models.Keyword`

        :param str name: keyword name
        :param str identifier: keyword identifier
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Keyword` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Keyword)

        model_queries_config = (
            (name, models.Keyword.name),
            (identifier, models.Keyword.identifier)
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_many_to_many_queries(q, ((entry_name, models.Keyword.entries, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def entry(self,
              name=None,
              dataset=None,
              recommended_full_name=None,
              recommended_short_name=None,
              gene_name=None,
              taxid=None,
              accession=None,
              organism_host=None,
              feature_type=None,
              function_=None,
              ec_number=None,
              db_reference=None,
              alternative_name=None,
              disease_comment=None,
              disease_name=None,
              tissue_specificity=None,
              pmid=None,
              keyword=None,
              subcellular_location=None,
              tissue_in_reference=None,
              sequence=None,
              limit=None,
              as_df=False):
        """Method to query :class:`pyuniprot.manager.Entry`

        An entry is the root element in UniProt datasets. Everything is linked to entry and can be accessed from
        :param dataset:
        :class:`models.Entry` object. `%` can be used as wildcard for string parameters (see examples below).

        .. seealso::

            :class:`pyuniprot.manager.models.Entry`

        :param str,tuple name: UniProt entry name(s)
        :param str,tuple recommended_full_name: recommended full protein name(s)
        :param str,tuple recommended_short_name: recommended short protein name(s)
        :param str,tuple tissue_in_reference: tissue mentioned in reference
        :param str,tuple subcellular_location: subcellular location(s)
        :param str,tuple keyword: keyword
        :param str,tuple pmid: PubMed identifier
        :param str,tuple tissue_specificity: tissue specificities
        :param str,tuple disease_comment: disease_comments
        :param str,tuple alternative_name:
        :param str,tuple db_reference: cross reference identifier
        :param str,tuple ec_number: enzyme classification number, e.g. 1.1.1.1
        :param str,tuple function_: description of protein functions
        :param str,tuple feature_type: feature types
        :param str,tuple organism_host: organism hosts
        :param str,tuple accession: UniProt accession number
        :param str,tuple disease_name: disease name
        :param str,tuple gene_name: gene name
        :param str,tuple taxid: NCBI taxonomy identifier
        :param int,tuple limit: maximum number of results
        :param str,tuple sequence: Amino acid sequence
        :param bool as_df: if set to True result returns as `pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Entry` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Entry)

        model_queries_config = (
            (dataset, models.Entry.dataset),
            (name, models.Entry.name),
            (recommended_full_name, models.Entry.recommended_full_name),
            (recommended_short_name, models.Entry.recommended_short_name),
            (gene_name, models.Entry.gene_name),
            (taxid, models.Entry.taxid),
        )
        q = self.get_model_queries(q, model_queries_config)

        one_to_many_queries_config = (
            (accession, models.Accession.accession),
            (organism_host, models.OrganismHost.taxid),
            (feature_type, models.Feature.type_),
            (function_, models.Function.text),
            (ec_number, models.ECNumber.ec_number),
            (db_reference, models.DbReference.identifier),
            (alternative_name, models.AlternativeFullName.name),
            (disease_comment, models.DiseaseComment.comment),
            (tissue_specificity, models.TissueSpecificity.comment),
            (sequence, models.Sequence.sequence),
        )
        q = self.get_one_to_many_queries(q, one_to_many_queries_config)

        many_to_many_queries_config = (
            (pmid, models.Entry.pmids, models.Pmid.pmid),
            (keyword, models.Entry.keywords, models.Keyword.name),
            (subcellular_location, models.Entry.subcellular_locations, models.SubcellularLocation.location),
            (tissue_in_reference, models.Entry.tissue_in_references, models.TissueInReference.tissue)
        )
        q = self.get_many_to_many_queries(q, many_to_many_queries_config)

        if disease_name:
            q = q.join(models.Entry.disease_comments).join(models.DiseaseComment.disease)
            if isinstance(disease_name, str):
                q = q.filter(models.Disease.name.like(disease_name))
            elif isinstance(disease_name, Iterable):
                q = q.filter(models.Disease.name.in_(disease_name))

        return self._limit_and_df(q, limit, as_df)

    def disease(self,
                identifier=None,
                ref_id=None,
                ref_type=None,
                name=None,
                acronym=None,
                description=None,
                entry_name=None,
                limit=None,
                as_df=False
                ):
        """Method to query :class:`pyuniprot.manager.models.Disease`

        .. seealso::

            :class:`pyuniprot.manager.models.Disease`

        :param identifier: disease UniProt identifier
        :param ref_id: identifier of referenced database
        :param ref_type: database name
        :param name: disease name
        :param acronym: disease acronym
        :param description: disease description
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Disease` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Disease)

        model_queries_config = (
            (identifier, models.Disease.identifier),
            (ref_id, models.Disease.ref_id),
            (ref_type, models.Disease.ref_type),
            (name, models.Disease.name),
            (acronym, models.Disease.acronym),
            (description, models.Disease.description)
        )
        q = self.get_model_queries(q, model_queries_config)

        if entry_name:
            q = q.session.query(models.Disease).join(models.DiseaseComment).join(models.Entry)
            if isinstance(entry_name, str):
                q = q.filter(models.Entry.name == entry_name)
            elif isinstance(entry_name, Iterable):
                q = q.filter(models.Entry.name.in_(entry_name))

        return self._limit_and_df(q, limit, as_df)

    def disease_comment(self, comment=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.models.DiseaseComment`

        .. seealso::

            :class:`pyuniprot.manager.models.DiseaseComment`

        :param comment: Comment to disease
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: Number of results, if limit=`None`, all results returned
        :param bool as_df: If `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.DiseaseComment` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.DiseaseComment)

        q = self.get_model_queries(q, ((comment, models.DiseaseComment.comment),))

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def other_gene_name(self, type_=None, name=None, entry_name=None, limit=None, as_df=None):
        """Method to query :class:`pyuniprot.manager.OtherGeneName`

        .. seealso::

            :class:`pyuniprot.manager.models.OtherGeneName`

        :param str type_: type of gene name e.g. *synonym*
        :param str name: other gene name
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: Number of results, if limit=`None`, all results returned
        :param bool as_df: If `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.DiseaseComment` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.OtherGeneName)

        model_queries_config = (
            (type_, models.OtherGeneName.type_),
            (name, models.OtherGeneName.name),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def alternative_full_name(self, name=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.AlternativeFullName`

        .. seealso::

            :class:`pyuniprot.manager.models.AlternativeFullName`

        :param str name: alternative full name
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.AlternativeFullName` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.AlternativeFullName)

        model_queries_config = (
            (name, models.AlternativeFullName.name),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def alternative_short_name(self, name=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.AlternativeShortlName`

        :param str name: alternative short name
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.AlternativeShortName` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.AlternativeShortName)

        model_queries_config = (
            (name, models.AlternativeShortName.name),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def accession(self, accession=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.Accession`

        :param str accession: UniProt Accession number
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Accession` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Accession)

        model_queries_config = (
            (accession, models.Accession.accession),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def pmid(self,
             pmid=None,
             entry_name=None,
             first=None,
             last=None,
             volume=None,
             name=None,
             date=None,
             title=None,
             limit=None,
             as_df=False
             ):
        """Method to query :class:`pyuniprot.manager.Pmid`

        .. seealso::

            :class:`pyuniprot.manager.models.Pmid`

        :param int pmid: PubMed identifier
        :param str entry_name: name in :class:`.models.Entry`
        :param first: first page
        :param last: last page
        :param volume: volume
        :param name: name of journal
        :param date: publication date
        :param title: title of publication
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Pmid` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Pmid)

        model_queries_config = (
            (pmid, models.Pmid.pmid),
            (last, models.Pmid.last),
            (first, models.Pmid.first),
            (volume, models.Pmid.volume),
            (name, models.Pmid.name),
            (date, models.Pmid.date),
            (title, models.Pmid.title)
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_many_to_many_queries(q, ((entry_name, models.Pmid.entries, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def organism_host(self, taxid=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.OrganismHost`

        .. seealso::

            :class:`pyuniprot.manager.models.OrganismHost`

        :param taxid: NCBI taxonomy identifier
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.OrganismHostt` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.OrganismHost)

        q = self.get_model_queries(q, ((taxid, models.OrganismHost.taxid),))

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def db_reference(self, type_=None, identifier=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.models.DbReference`

        Check list of available databases with on :py:attr:`.dbreference_types`

        .. seealso::

            :class:`pyuniprot.manager.models.DbReference`

        :param type_: type (or name) of database
        :param identifier: unique identifier in database
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.DbReference` objects or :class:`pandas.DataFrame`

        **Links**

        - `UniProt dbxref <http://www.uniprot.org/docs/dbxref>`_
        """
        q = self.session.query(models.DbReference)

        model_queries_config = (
            (type_, models.DbReference.type_),
            (identifier, models.DbReference.identifier)
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def feature(self, type_=None, identifier=None, description=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.Feature`

        Check available features types with ``pyuniprot.query().feature_types``

        .. seealso::

            :class:`pyuniprot.manager.models.Feature`

        :param type_: type of feature
        :param identifier: feature identifier
        :param description: description of feature
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Feature` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Feature)

        model_queries_config = (
            (type_, models.Feature.type_),
            (identifier, models.Feature.identifier),
            (description, models.Feature.description)
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def function(self, text=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.Function`

        .. seealso::

            :class:`pyuniprot.manager.models.Function`

        :param text: description of function
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.Function` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Function)

        model_queries_config = (
            (text, models.Function.text),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def ec_number(self, ec_number=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.ECNumber`

        .. seealso::

            :class:`pyuniprot.manager.models.ECNumber`

        :param ec_number: Enzyme Commission number
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.ECNumber` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.ECNumber)

        q = self.get_model_queries(q, ((ec_number, models.ECNumber.ec_number),))

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def sequence(self, sequence=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.Sequence`

        .. seealso::

            :class:`pyuniprot.manager.models.Sequence`

        :param sequence: AA sequence
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.SubcellularLocation` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.Sequence)

        q = self.get_model_queries(q, ((sequence, models.Sequence.sequence),))

        q = self.get_many_to_many_queries(q, ((entry_name, models.SubcellularLocation.entries, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def subcellular_location(self, location=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.SubcellularLocation`

        .. seealso::

            :class:`pyuniprot.manager.models.SubcellularLocation`

        :param location: subcellular location
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.SubcellularLocation` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.SubcellularLocation)

        q = self.get_model_queries(q, ((location, models.SubcellularLocation.location),))

        q = self.get_many_to_many_queries(q, ((entry_name, models.SubcellularLocation.entries, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def tissue_specificity(self, comment=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.TissueSpecificity`

        Provides information on the expression of a gene at the mRNA or protein level in cells or in tissues of
        multicellular organisms. By default, the information is derived from experiments at the mRNA level, unless
        specified ‘at protein level

        .. seealso::

            :class:`pyuniprot.manager.models.TissueSpecificity`

        :param str comment: Comment describing tissue specificity
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if `None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`pyuniprot.manager.models.TissueSpecificity` objects or :class:`pandas.DataFrame`
        """
        q = self.session.query(models.TissueSpecificity)

        q = self.get_model_queries(q, ((comment, models.TissueSpecificity.comment),))

        q = self.get_one_to_many_queries(q, ((entry_name, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    def tissue_in_reference(self, tissue=None, entry_name=None, limit=None, as_df=False):
        """Method to query :class:`pyuniprot.manager.TissueInReference`

        :param str tissue: tissue linked to reference
        :param str entry_name: name in :class:`.models.Entry`
        :param int,tuple limit: number of results, if limit=`None`, all results returned
        :param bool as_df: if `True` results are returned as :class:`pandas.DataFrame`
        :return: list of :class:`models.TissueInReference` objects or :class:`pandas.DataFrame`
        :rtype: [models.TissueInReference,] or [pandas.DataFrame]
        """
        q = self.session.query(models.TissueInReference)

        model_queries_config = (
            (tissue, models.TissueInReference.tissue),
        )
        q = self.get_model_queries(q, model_queries_config)

        q = self.get_many_to_many_queries(q, ((entry_name, models.TissueInReference.entries, models.Entry.name),))

        return self._limit_and_df(q, limit, as_df)

    @property
    def dbreference_types(self):
        """Distinct database reference types (``type_``) in :class:`pyuniprot.manager.models.DbReference`

        :return: List of strings for all available database cross reference types used in model DbReference
        :rtype: [str,]
        """
        q = self.session.query(distinct(models.DbReference.type_))
        return [x[0] for x in q.all()]

    @property
    def taxids(self):
        """Distinct NCBI taxonomy identifiers (``taxid``) in :class:`pyuniprot.manager.models.Entry`

        :return: NCBI taxonomy identifiers
        :rtype: [int,]
        """
        r = self.session.query(distinct(models.Entry.taxid)).all()
        return [x[0] for x in r]

    @property
    def datasets(self):
        """Distinct datasets (``dataset``) in :class:`pyuniprot.manager.models.Entry`

        Distinct datasets are SwissProt or/and TrEMBL

        :return: all distinct dataset types
        :rtype: [str,]
        """
        r = self.session.query(distinct(models.Entry.dataset)).all()
        return [x[0] for x in r]

    @property
    def feature_types(self):
        """Distinct types (``type_``) in :class:`pyuniprot.manager.models.Feature`

        :return: all distinct feature types
        :rtype: [str,]
        """
        r = self.session.query(distinct(models.Feature.type_)).all()
        return [x[0] for x in r]

    @property
    def subcellular_locations(self):
        """Distinct subcellular locations (``location`` in :class:`pyuniprot.manager.models.SubcellularLocation`)

        :return: all distinct subcellular locations
        :rtype: [str,]
        """
        return [x[0] for x in self.session.query(models.SubcellularLocation.location).all()]

    @property
    def tissues_in_references(self):
        """Distinct tissues (``tissue`` in :class:`pyuniprot.manager.models.TissueInReference`)

        :return: all distinct tissues in references
        :rtype: [str,]
        """
        return [x[0] for x in self.session.query(models.TissueInReference.tissue).all()]

    @property
    def keywords(self):
        """Distinct keywords (``name`` in :class:`pyuniprot.manager.models.Keyword`)

        :returns: all distinct keywords
        :rtype: [str,]
        """
        return [x[0] for x in self.session.query(models.Keyword.name).all()]

    @property
    def diseases(self):
        """Distinct diseases (``name`` in :class:`pyuniprot.manager.models.Disease`)

        :returns: all distinct disease names
        :rtype: [str,]
        """
        return [x[0] for x in self.session.query(models.Disease.name).all()]

    @property
    def version(self):
        """Version of UniPort knowledgebase

        :returns: dictionary with version info
        :rtype: dict
        """
        return [x for x in self.session.query(models.Version).all()]