:py:mod:`cobra.util.solver` =========================== .. py:module:: cobra.util.solver .. autoapi-nested-parse:: Additional helper functions for the optlang solvers. All functions integrate well with the context manager, meaning that all operations defined here are automatically reverted when used in a `with model:` block. The functions defined here together with the existing model functions should allow you to implement custom flux analysis methods with ease. Module Contents --------------- Classes ~~~~~~~ .. autoapisummary:: cobra.util.solver.Components Functions ~~~~~~~~~ .. autoapisummary:: cobra.util.solver.linear_reaction_coefficients cobra.util.solver._valid_atoms cobra.util.solver.set_objective cobra.util.solver.interface_to_str cobra.util.solver.get_solver_name cobra.util.solver.choose_solver cobra.util.solver.check_solver cobra.util.solver.add_cons_vars_to_problem cobra.util.solver.remove_cons_vars_from_problem cobra.util.solver.add_absolute_expression cobra.util.solver.fix_objective_as_constraint cobra.util.solver.check_solver_status cobra.util.solver.assert_optimal cobra.util.solver.add_lp_feasibility cobra.util.solver.add_lexicographic_constraints Attributes ~~~~~~~~~~ .. autoapisummary:: cobra.util.solver.CONS_VARS cobra.util.solver.logger cobra.util.solver.solvers cobra.util.solver.qp_solvers cobra.util.solver.has_primals .. py:data:: CONS_VARS .. py:data:: logger .. py:data:: solvers .. py:data:: qp_solvers :value: ['cplex', 'gurobi', 'osqp'] .. py:data:: has_primals .. py:class:: Components Bases: :py:obj:`NamedTuple` Define an object for adding absolute expressions. .. py:attribute:: variable :type: optlang.interface.Variable .. py:attribute:: upper_constraint :type: optlang.interface.Constraint .. py:attribute:: lower_constraint :type: optlang.interface.Constraint .. py:function:: linear_reaction_coefficients(model: cobra.Model, reactions: Optional[List[cobra.Reaction]] = None) -> Dict[cobra.Reaction, float] Retrieve coefficient for the reactions in a linear objective. :param model: The cobra model defining the linear objective. :type model: cobra.Model :param reactions: An optional list of the reactions to get the coefficients for. By default, all reactions are considered (default None). :type reactions: list of cobra.Reaction, optional :returns: A dictionary where the keys are the reaction objects and the values are the corresponding coefficient. Empty dictionary if there are no linear terms in the objective. :rtype: dict .. py:function:: _valid_atoms(model: cobra.Model, expression: optlang.symbolics.Basic) -> bool Check whether a sympy expression references the correct variables. :param model: The model in which to check for variables. :type model: cobra.Model :param expression: A sympy expression. :type expression: sympy.Basic :returns: True if all referenced variables are contained in model, False otherwise. :rtype: bool .. py:function:: set_objective(model: cobra.Model, value: Union[optlang.interface.Objective, optlang.symbolics.Basic, Dict[cobra.Reaction, float]], additive: bool = False) -> None Set the model objective. :param model: The model to set the objective for. :type model: cobra.Model :param value: If the model objective is linear, then the value can be a new optlang.interface.Objective or a dictionary with linear coefficients where each key is a reaction and the corresponding value is the new coefficient (float). If the objective is non-linear and `additive` is True, then only values of class optlang.interface.Objective, are accepted. :type value: optlang.interface.Objective, optlang.symbolics.Basic, dict :param additive: If True, add the terms to the current objective, otherwise start with an empty objective. :type additive: bool :raises ValueError: If model objective is non-linear and the `value` is a dict. :raises TypeError: If the type of `value` is not one of the accepted ones. .. py:function:: interface_to_str(interface: Union[str, types.ModuleType]) -> str Give a string representation for an optlang interface. :param interface: Full name of the interface in optlang or cobra representation. For instance, 'optlang.glpk_interface' or 'optlang-glpk'. :type interface: str, ModuleType :returns: The name of the interface as a string. :rtype: str .. py:function:: get_solver_name(mip: bool = False, qp: bool = False) -> str Select a solver for a given optimization problem. :param mip: True if the solver requires mixed integer linear programming capabilities. :type mip: bool :param qp: True if the solver requires quadratic programming capabilities. :type qp: bool :returns: The name of the feasible solver. :rtype: str :raises SolverNotFound: If no suitable solver could be found. .. py:function:: choose_solver(model: cobra.Model, solver: Optional[str] = None, qp: bool = False) -> types.ModuleType Choose a solver given a solver name and model. This will choose a solver compatible with the model and required capabilities. Also respects model.solver where it can. :param model: The model for which to choose the solver. :type model: cobra.Model :param solver: The name of the solver to be used (default None). :type solver: str, optional :param qp: True if the solver needs quadratic programming capabilities (default False). :type qp: boolean, optional :returns: Valid solver for the problem. :rtype: optlang.interface :raises SolverNotFound: If no suitable solver could be found. .. py:function:: check_solver(obj) Check whether the chosen solver is valid. Check whether chosen solver is valid and also warn when using a specialized solver. Will return the optlang interface for the requested solver. :param obj: The chosen solver. :type obj: str or optlang.interface or optlang.interface.Model :raises SolverNotFound: If the solver is not valid. .. py:function:: add_cons_vars_to_problem(model: cobra.Model, what: Union[List[CONS_VARS], Tuple[CONS_VARS], Components], **kwargs) -> None Add variables and constraints to a model's solver object. Useful for variables and constraints that can not be expressed with reactions and lower/upper bounds. It will integrate with the model's context manager in order to revert changes upon leaving the context. :param model: The model to which to add the variables and constraints. :type model: cobra.Model :param what: optlang.interface.Constraint The variables and constraints to add to the model. :type what: list or tuple of optlang.interface.Variable or :param \*\*kwargs: Keyword arguments passed to solver's add() method. :type \*\*kwargs: keyword arguments .. py:function:: remove_cons_vars_from_problem(model: cobra.Model, what: Union[List[CONS_VARS], Tuple[CONS_VARS], Components]) -> None Remove variables and constraints from a model's solver object. Useful to temporarily remove variables and constraints from a model's solver object. :param model: The model from which to remove the variables and constraints. :type model: cobra.Model :param what: optlang.interface.Constraint The variables and constraints to remove from the model. :type what: list or tuple of optlang.interface.Variable or .. py:function:: add_absolute_expression(model: cobra.Model, expression: str, name: str = 'abs_var', ub: Optional[float] = None, difference: float = 0.0, add: bool = True) -> Components Add the absolute value of an expression to the model. Also defines a variable for the absolute value that can be used in other objectives or constraints. :param model: The model to which to add the absolute expression. :type model: cobra.Model :param expression: Must be a valid symbolic expression within the model's solver object. The absolute value is applied automatically on the expression. :type expression: str :param name: The name of the newly created variable (default "abs_var"). :type name: str, optional :param ub: The upper bound for the variable (default None). :type ub: positive float, optional :param difference: The difference between the expression and the variable (default 0.0). :type difference: positive float, optional :param add: Whether to add the variable to the model at once (default True). :type add: bool, optional :returns: A named tuple with variable and two constraints (upper_constraint, lower_constraint) describing the new variable and the constraints that assign the absolute value of the expression to it. :rtype: Components .. py:function:: fix_objective_as_constraint(model: cobra.Model, fraction: float = 1.0, bound: Optional[float] = None, name: str = 'fixed_objective_{}') -> float Fix current objective as an additional constraint. When adding constraints to a model, such as done in pFBA which minimizes total flux, these constraints can become too powerful, resulting in solutions that satisfy optimality but sacrifices too much for the original objective function. To avoid that, we can fix the current objective value as a constraint to ignore solutions that give a lower (or higher depending on the optimization direction) objective value than the original model. When done with the model as a context, the modification to the objective will be reverted when exiting that context. :param model: The model to operate on. :type model: cobra.Model :param fraction: The fraction of the optimum the objective is allowed to reach (default 1.0). :type fraction: float, optional :param bound: The bound to use instead of fraction of maximum optimal value. If not None, `fraction` is ignored (default None). :type bound: float, optional :param name: Name of the objective. May contain one "{}" placeholder which is filled with the name of the old objective (default "fixed_objective_{}"). :type name: str, optional :returns: The value of the optimized objective * fraction :rtype: float .. py:function:: check_solver_status(status: str = None, raise_error: bool = False) -> None Perform standard checks on a solver's status. :param status: The status string obtained from the solver (default None). :type status: str, optional :param raise_error: If True, raise error or display warning if False (default False). :type raise_error: bool, optional :rtype: None :Warns: **UserWarning** -- If `status` is not optimal and `raise_error` is set to True. :raises OptimizationError: If `status` is None or is not optimal and `raise_error` is set to True. .. py:function:: assert_optimal(model: cobra.Model, message: str = 'Optimization failed') -> None Assert model solver status is optimal. Do nothing if model solver status is optimal, otherwise throw appropriate exception depending on the status. :param model: The model to check the solver status for. :type model: cobra.Model :param message: Message for the exception if solver status is not optimal (default "Optimization failed"). :type message: str, optional :rtype: None :raises OptimizationError: If solver status is not optimal. .. py:function:: add_lp_feasibility(model: cobra.Model) -> None Add a new objective and variables to ensure a feasible solution. The optimized objective will be zero for a feasible solution and otherwise represent the distance from feasibility (please see [1]_ for more information). :param model: The model whose feasibility is to be tested. :type model: cobra.Model :rtype: None .. rubric:: References .. [1] Gomez, Jose A., Kai Höffner, and Paul I. Barton. “DFBAlab: A Fast and Reliable MATLAB Code for Dynamic Flux Balance Analysis.” BMC Bioinformatics 15, no. 1 (December 18, 2014): 409. https://doi.org/10.1186/s12859-014-0409-8. .. py:function:: add_lexicographic_constraints(model: cobra.Model, objectives: List[cobra.Reaction], objective_direction: Union[str, List[str]] = 'max') -> pandas.Series Successively optimize separate targets in a specific order. For each objective, optimize the model and set the optimal value as a constraint. Proceed in the order of the objectives given. Due to the specific order this is called lexicographic FBA [1]_. This procedure is useful for returning unique solutions for a set of important fluxes. Typically this is applied to exchange fluxes. :param model: The model to be optimized. :type model: cobra.Model :param objectives: A list of reactions (or objectives) in the model for which unique fluxes are to be determined. :type objectives: list of cobra.Reaction :param objective_direction: The desired objective direction for each reaction (if a list) or the objective direction to use for all reactions (default "max"). :type objective_direction: str or list of str, optional :returns: A pandas Series containing the optimized fluxes for each of the given reactions in `objectives`. :rtype: pandas.Series .. rubric:: References .. [1] Gomez, Jose A., Kai Höffner, and Paul I. Barton. “DFBAlab: A Fast and Reliable MATLAB Code for Dynamic Flux Balance Analysis.” BMC Bioinformatics 15, no. 1 (December 18, 2014): 409. https://doi.org/10.1186/s12859-014-0409-8.