`rbatools.rba_problem` module

Classes

class ProblemRBA

Class holding RBA-problem as mathematical manifestation of the RBA-model.

Attributes

classicRBA : boolean: Indicates that the problem is a classic RBA-problem (as defined in RBApy)
ClassicRBAmatrix : rba.ConstraintMatrix.matrix object
Mu : float: Current Growth rate as numeric value
LP : LinearProblem object
Enzyme_FWcapacities : list: List of constraint IDs, which represent forward efficiencies of enzymes Created by method 'extract_constraint_types'
Enzyme_BWcapacities : list: List of constraint IDs, which represent backward efficiencies of enzymes Created by method 'extract_constraint_types'
ProcessCapacities : list: List of constraint IDs, which represent net efficiencies of processes Created by method 'extract_constraint_types'
Metabolites : list: List of constraint IDs, which represent mass balances of metabolites Created by method 'extract_constraint_types'
CompartmentDensities : list: List of constraint IDs, which represent compartment capacities Created by method 'extract_constraint_types'
Reactions : list: List of constraint IDs, which represent metabolic reactions Created by method 'extract_variable_types'
Enzymes : list: List of constraint IDs, which represent enzymes, associated with metabolic reactions Created by method 'extract_variable_types'
Processes : list: List of constraint IDs, which represent process machineries Created by method 'extract_variable_types'
MuDepIndices_A : list: List of tuples holding rows (constraint IDs) and columns (variable IDs) of constraint-matrix coefficients (LHS), which depend on the growth rate. Created by method 'find_growth_rate_dependencies'
MuDepIndices_b : list: List of constraint IDs, whos RHS depend on the growth rate Created by method 'find_growth_rate_dependencies'
MuDepIndices_LB : list: List of variable IDs, whos lower-bounds depend on the growth rate Created by method 'find_growth_rate_dependencies'
MuDepIndices_UB : list: List of variable IDs, whos upper-bounds depend on the growth rate Created by method 'find_growth_rate_dependencies'
Solved : bool: Booelean indicating wheter Problem has been successfully solved, according to the acceptable solution statuses provided to the solve_lp-method. Created by method 'solve_lp'
SolutionStatus : int: Numerical value indicating the solution-status of the problem. Consult CPLEX documentation for their meaning. Created by method 'solve_lp'
ObjectiveValue : float: Numeric value of the objective function after optimisation by CPLEX. Created by method 'solve_lp'
SolutionValues : dict: Solution vector after optimisation by CPLEX. (Dictionary with variable IDs as keys and numeric values as values) Created by method 'solve_lp'
DualValues : dict: Vector of dual-values after optimisation by CPLEX. (Dictionary with constraint IDs as keys and numeric values as values) Created by method 'solve_lp'

Methods

def __init__(self, matrix, ModelStructure, lp_solver='cplex')

Initiates rbatools.rba_problem.ProblemRBA from supplied rba.ConstraintMatrix object.

Makes sure the problem is consistent and optimisable.

Parameters

matrix : rba.ConstraintMatrix
ModelStructure rbatools.rba_model_structure.ModelStructureRBA
lp_solver : str: Specifies which LP-solver should be used ('cplex' or 'swiglpk') Default: 'cplex'

def build_classic_matrix(self, Mu)

Builds standard RBA-matrix according to growth-rate

Parameters

Mu : float: Growth rate

def calculate_left_hand_side(self, constraints=[])

Calculates value of problem's lefthand side (after multiplying with solution-vector).

Parameters

constraints : str or list of str: Constraints to retreive the LHS-value for. Either constraint ID or list of constraint IDs to specify the LHS-value of which constraint to look up. This is an optional input; if not provided all constraint are looked up.

Returns

dict: Dictionary with constraint-IDs as keys and LHS-value as values.

def clear_objective(self)

Sets all coefficients of the objective function to zero.

def evaluate_parameter(self, definition)

def extract_constraint_types(self, ModelStructure)

Extracts information on the different constraint types in the standard RBA-matrix.

AddsAttributes : Enzyme_FWcapacities, Enzyme_BWcapacities, ProcessCapacities,: Metabolites and CompartmentDensities

def extract_variable_types(self, ModelStructure)

Extracts information on the different variable types in the standard RBA-matrix.

AddsAttributes : Enzymes, Reactions and Processes

def find_growth_rate_dependencies(self, ModelStructure)

Extracts information on the growth-rate dependent LP-coefficients.

AddsAttributes : MuDepIndices_A, MuDepIndices_b, MuDepIndices_LB: and MuDepIndices_UB

def get_constraint_types(self, constraints=[])

Extracts type of constraints.

Parameters

constraints : str or list of str: Constraints to retreive the type for. Either constraint ID or list of constraint IDs to specify the type of which constraint to look up. This is an optional input; if not provided all constraint are looked up.

Returns

dict: Dictionary with constraint-IDs as keys and type identification-characters as values.

def get_lb(self, variables=[])

Returns lower bounds of problem variables.

Parameters

variables : str list of str: Variables to retreive the objective coefficients for. Either variable ID or list of variable IDs to specify the coefficients of which variables to look up. This is an optional input; if not provided all variables are looked up.

Returns

dict: Dictionary with variable-IDs as keys and lower bounds as values.

def get_objective(self, variables=[])

Returns objective coefficient of variables in problem.

Parameters

variables : str or list of str: Variables to retreive the objective coefficients for. Either variable ID or list of variable IDs to specify the coefficients of which variables to look up. This is an optional input; if not provided all variables are looked up.

Returns

dict: Dictionary with variable-IDs as keys and objective coefficients as values.

def get_problem_coefficients(self, inputTuples=[])

Returns coefficients of LHS of problem.

Parameters

inputTuples : tuple or list of tuples. Tuples hold row and column indices. [('row1','col1'),('row2','col2'),…] or ('row1','col1').

Returns

dict: Dictionary with index tuples as keys and matrix coefficients as values.

def get_right_hand_side(self, constraints=[])

Extracts coefficients of problem's righthand side (B-vector).

Parameters

constraints : str or list of str: Constraints to retreive the objective coefficients for. Either constraint ID or list of constraint IDs to specify the RHS of which constraint to look up. This is an optional input; if not provided all constraint are looked up.

Returns

dict: Dictionary with constraint-IDs as keys and RHS-values as values.

def get_ub(self, variables=[])

Returns upper bounds of problem variables.

Parameters

variables : str list of str: Variables to retreive the objective coefficients for. Either variable ID or list of variable IDs to specify the coefficients of which variables to look up. This is an optional input; if not provided all variables are looked up.

Returns

dict: Dictionary with variable-IDs as keys and upper bounds as values.

def invert_objective(self)

Changes sign (optimisation-sense) of objective function.

def set_constraint_types(self, inputDict)

Sets type of constraints.

E : = ; L: <= ; G: >=

Parameters

inputDict : dict: Dictionary with constraint-IDs as keys and type identification-character as values. ({'col1':'E','col2':'L', …}).

def set_growth_rate(self, Mu, ModelStructure, keepParameters={})

Changes growth-rate of problem and sets all associated coefficients.

Can be provided with 'keepParameters' argument to define problem coefficients which should remain unchanged.

Parameters

Mu : float: Growth rate
ModelStructure : RBA_ModellStructure object.
keepParameters : dict: Dictionary indicating which elements of the linear problem should not be affected when setting growth-rate. Possible keys of dictionary: 'LHS': List of index-tuples (constraint-ID,variable-ID), indicating elements of the lefthandside (constraint-matrix). 'RHS': List of constraint IDs, indicating elements of the righthandside (b-vector). 'LB': List of variable-IDs, indicating elements of the lower-bound vector. 'UB': List of variable-IDs, indicating elements of the upper-bound vector. Default: keepParameters=None

def set_lb(self, inputDict)

Set lower-bounds of the problem variables.

Parameters

inputDict : dict: Dictionary with variable-IDs as keys and new numeric values as values. ({'col1':42,'col2':9000, …}).

def set_objective(self, inputDict)

Sets objective function coefficients.

Parameters

inputDict : dict: Dictionary with variable-IDs as keys and new numeric values as values. ({'col1':42,'col2':9000, …}).

def set_problem_coefficients(self, inputDict)

Set coefficients of the problems' LHS (constraint matrix).

Parameters

inputDict : dict: Dict with index-tuples ('row1','col1') as keys and new numeric values as values. ({('row1','col1'):42,('row2','col2'):9000, …}).

def set_right_hand_side(self, inputDict)

Set coefficients of the problems' RHS (b-vector). Parameters

inputDict : dict: Dictionary with constraint-IDs as keys and new numeric values as values. ({'row1':42,'row2':9000, …}).

def set_ub(self, inputDict)

Set upper-bounds of the problem variables.

Parameters

inputDict : dict: Dictionary with variable-IDs as keys and new numeric values as values. ({'col1':42,'col2':9000, …}).

def solve_lp(self, feasible_stati=['optimal', 'feasible'], try_unscaling_if_sol_status_is_feasible_only_before_unscaling=True)

Solves Linear RBA problem.

When solution-status is amongst the user-defined feasible statuses; boolean 'Solved' is set to True and 'ObjectiveValue', 'SolutionValues' and 'DualValues' are stored as attributes.

Parameters

feasible_stati : list of int: List with identifiers of acceptable solution statuses. (consult ILOG-CPLEX documentation for information on them). Default: feasible_stati=["optimal","feasible"]

try_unscaling_if_sol_status_is_feasible_only_before_unscaling :bool Try re-solving problem with less aggressive scaling, when solution status is infeasible after unscaling Default: True

def update_growth_rate(self, Mu, keepParameters={}, ModifiedProblem=False)

Changes growth-rate of problem and sets all associated coefficients.

Can be provided with 'keepParameters' argument to define problem coefficients which should remain unchanged.

Parameters

Mu : float: Growth rate
keepParameters : dict: Dictionary indicating which elements of the linear problem should not be affected when setting growth-rate. Possible keys of dictionary: 'LHS': List of index-tuples (constraint-ID,variable-ID), indicating elements of the lefthandside (constraint-matrix). 'RHS': List of constraint IDs, indicating elements of the righthandside (b-vector). 'LB': List of variable-IDs, indicating elements of the lower-bound vector. 'UB': List of variable-IDs, indicating elements of the upper-bound vector. Default: keepParameters=None
ModifiedProblem : bool: Default: False