DigestionEnzymeDB.h

Go to the documentation of this file.

// Copyright (c) 2002-present, OpenMS Inc. -- EKU Tuebingen, ETH Zurich, and FU Berlin
// SPDX-License-Identifier: BSD-3-Clause
//
// --------------------------------------------------------------------------
// $Maintainer: Xiao Liang $
// $Authors: Xiao Liang, Chris Bielow $
// --------------------------------------------------------------------------
 
#pragma once
 
#include <OpenMS/CHEMISTRY/DigestionEnzyme.h>
#include <OpenMS/CONCEPT/Exception.h>
#include <OpenMS/CONCEPT/LogStream.h>
#include <OpenMS/DATASTRUCTURES/String.h>
#include <OpenMS/FORMAT/ParamXMLFile.h>
#include <OpenMS/SYSTEM/File.h>
 
#include <set>
#include <map>
 
namespace OpenMS
{
  template<typename DigestionEnzymeType, typename InstanceType> class DigestionEnzymeDB
  {
  public:
 
    typedef typename std::set<const DigestionEnzymeType*>::const_iterator ConstEnzymeIterator;
    typedef typename std::set<const DigestionEnzymeType*>::iterator EnzymeIterator;
 
    static InstanceType* getInstance()
    {
      static InstanceType* db_ = nullptr;
      if (db_ == nullptr)
      {
        db_ = new InstanceType;
      }
      return db_;
    }
 
    virtual ~DigestionEnzymeDB()
    {
      for (ConstEnzymeIterator it = const_enzymes_.begin(); it != const_enzymes_.end(); ++it)
      {
        delete *it;
      }
    }
 
    const DigestionEnzymeType* getEnzyme(const String& name) const
    {
      auto pos = enzyme_names_.find(name);
      if (pos == enzyme_names_.end())
      {
        throw Exception::ElementNotFound(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, name);
      }
      return pos->second;
    }
 
    const DigestionEnzymeType* getEnzymeByRegEx(const String& cleavage_regex) const
    {
      if (!hasRegEx(cleavage_regex))
      {
        // @TODO: why does this use a different exception than "getEnzyme"?
        throw Exception::IllegalArgument(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION,
                                         String("Enzyme with regex " + cleavage_regex + " was not registered in Enzyme DB, register first!").c_str());
      }
      return enzyme_regex_.at(cleavage_regex);
    }
 
    void getAllNames(std::vector<String>& all_names) const
    {
      all_names.clear();
      for (ConstEnzymeIterator it = const_enzymes_.begin(); it != const_enzymes_.end(); ++it)
      {
        all_names.push_back((*it)->getName());
      }
    }
 
    bool hasEnzyme(const String& name) const
    {
      return (enzyme_names_.find(name) != enzyme_names_.end());
    }
 
    bool hasRegEx(const String& cleavage_regex) const
    {
      return (enzyme_regex_.find(cleavage_regex) != enzyme_regex_.end());
    }
 
    bool hasEnzyme(const DigestionEnzymeType* enzyme) const
    {
      return (const_enzymes_.find(enzyme) != const_enzymes_.end() );
    }
 
    inline ConstEnzymeIterator beginEnzyme() const { return const_enzymes_.begin(); }  // we only allow constant iterators -- this DB is not meant to be modifiable
    inline ConstEnzymeIterator endEnzyme() const { return const_enzymes_.end(); }
 
  protected:
    DigestionEnzymeDB(const String& db_file = "")
    {
      if (!db_file.empty())
      {
        readEnzymesFromFile_(db_file);
      }
    }
 
    DigestionEnzymeDB(const DigestionEnzymeDB& enzymes_db) = delete;
 
    DigestionEnzymeDB& operator=(const DigestionEnzymeDB& enzymes_db) = delete;
 
    bool readEnzymesFromFileIfPresent_(const String& filename)
    {
      try
      {
        readEnzymesFromFile_(filename);
        return true;
      }
      catch (Exception::FileNotFound&)
      {
        // file not found - that's OK, we will use built-in enzymes
        return false;
      }
    }
 
    void readEnzymesFromFile_(const String& filename)
    {
      String file = File::find(filename);
 
      Param param;
      ParamXMLFile().load(file, param);
      if (param.empty()) return;
 
      std::vector<String> split;
      String(param.begin().getName()).split(':', split);
      if (split[0] != "Enzymes")
      {
        throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, split[0], "name 'Enzymes' expected");
      }
 
      try
      {
        std::map<String, String> values;
        String previous_enzyme = split[1];
        // this iterates over all the "ITEM" elements in the XML file:
        for (Param::ParamIterator it = param.begin(); it != param.end(); ++it)
        {
          String(it.getName()).split(':', split);
          if (split[0] != "Enzymes") break; // unexpected content in the XML file
          if (split[1] != previous_enzyme)
          {
            // add enzyme and reset:
            addEnzyme_(parseEnzyme_(values));
            previous_enzyme = split[1];
            values.clear();
          }
          values[it.getName()] = String(it->value.toString());
        }
        // add last enzyme
        addEnzyme_(parseEnzyme_(values));
      }
      catch (Exception::BaseException& e)
      {
        throw Exception::ParseError(__FILE__, __LINE__, OPENMS_PRETTY_FUNCTION, e.what(), "");
      }
    }
 
    const DigestionEnzymeType* parseEnzyme_(std::map<String, String>& values) const
    {
      DigestionEnzymeType* enzy_ptr = new DigestionEnzymeType();
 
      for (std::map<String, String>::iterator it = values.begin(); it != values.end(); ++it)
      {
        const String& key = it->first;
        const String& value = it->second;
        if (!enzy_ptr->setValueFromFile(key, value))
        {
          OPENMS_LOG_ERROR << "Error while parsing enzymes file: unknown key '" << key << "' with value '" << value << "'" << std::endl;
        }
      }
      return enzy_ptr;
    }
 
    void addEnzyme_(const DigestionEnzymeType* enzyme)
    {
      String name = enzyme->getName();
 
      // if an enzyme with the same name exists, remove the old one first
      auto existing = enzyme_names_.find(name);
      if (existing != enzyme_names_.end())
      {
        const DigestionEnzymeType* old = existing->second;
        const_enzymes_.erase(old);
        // remove old name/synonym entries
        String old_name = old->getName();
        enzyme_names_.erase(old_name);
        enzyme_names_.erase(old_name.toLower());
        for (const auto& syn : old->getSynonyms())
        {
          enzyme_names_.erase(syn);
        }
        // remove old regex entry
        if (!old->getRegEx().empty())
        {
          enzyme_regex_.erase(old->getRegEx());
        }
        delete old;
      }
 
      // add to internal storage
      const_enzymes_.insert(enzyme);
      // add to internal indices (by name and its synonyms)
      enzyme_names_[name] = enzyme;
      enzyme_names_[name.toLower()] = enzyme;
      for (std::set<String>::const_iterator it = enzyme->getSynonyms().begin(); it != enzyme->getSynonyms().end(); ++it)
      {
        enzyme_names_[*it] = enzyme;
      }
      // ... and by regex
      if (enzyme->getRegEx() != "")
      {
        enzyme_regex_[enzyme->getRegEx()] = enzyme;
      }
      return;
    }
 
    std::map<String, const DigestionEnzymeType*> enzyme_names_; 
 
    std::map<String, const DigestionEnzymeType*> enzyme_regex_; 
 
    std::set<const DigestionEnzymeType*> const_enzymes_; 
 
  };
}