MFE_calculator

Attributes

_PAIRS

Classes

RNAFolder

An RNA folding algorithm and parameters

RNAFolderBase

An RNA folding algorithm and parameters

RNAFolderBase2

An RNA folding algorithm and parameters

RNASolution

An RNA folding solution and working data

RNASolutionBase

An RNA folding solution and working data

RNASolutionBase2

An RNA folding solution and working data

Module Contents

class MFE_calculator.RNAFolder(loop_min=0, energies=None)

Bases: RNAFolderBase2

An RNA folding algorithm and parameters

An instance holds all the configuration necessary to solve an RNA folding problem. Note that energies are the free-energy (a more negative value for the more favourable pairings).

solve(rna)
Parameters:

rna (str)

class MFE_calculator.RNAFolderBase(loop_min=0, energies=None)

An RNA folding algorithm and parameters

An instance holds all the configuration necessary to solve an RNA folding problem. Note that energies are the free-energy (a more negative value for the more favourable pairings).

MAX_BULGE = 20
MAX_HAIRPIN = 20
MAX_INTERNAL = 20
static _helix_lookups(energies)

Create look-up tables for free energy in helices

Parameters:

energies (dict)

static _interpolated_table(energies, size, key, fmt='02d')

Read values key01, key02, … to array interpolating missing values.

We use this to encode e.g. the hairpin loop energies in a table we can use incrementally during minimisation.

Parameters:

  • energies (dict)

  • size (int)

  • key (str)

  • fmt (str)

Return type:

numpy.ndarray

static _loop_lookups(energies)

Create look-up tables for loop free energy in loops

Parameters:

energies (dict)

static _special_lookups(energies, dflt=None)

Get misc special free energy parameters

Parameters:

energies (dict)

lmin = 0
non_pairing = None
print_hairpin_fe(n)
Parameters:

n (int)

print_stack_fe(rp, sp)
Parameters:

  • rp (str)

  • sp (str)

print_term_fe(rp, an, bn)
Parameters:

  • rp (str)

  • an (str)

  • bn (str)

class MFE_calculator.RNAFolderBase2(loop_min=0, energies=None)

Bases: RNAFolderBase

An RNA folding algorithm and parameters

An instance holds all the configuration necessary to solve an RNA folding problem. Note that energies are the free-energy (a more negative value for the more favourable pairings).

closed_energy(soln, i, j)

Minimum energy and mode of [i, j] closed by pairing (i,j).

If the pair is forbidden (not AU, CG, GC, GU, UA or UG) the method returns the self.non_pairing energy (normally math.inf), and the returned mode is -1. Otherwise, the method compares the minimum energy obtained by pairing (i,j) across [i+1, j-1] considered as open and closed, and returns the lower energy and mode (0=open, 1=closed).

Parameters:
Return type:

tuple[float, int]

open_energy(soln, i, j)

Minimum energy of [i, j] without pairing (i,j).

The method seeks a lowest energy for [i,j], considering j unpaired or paired with any interior nucleotide k, where i<k<j.

Parameters:
Return type:

tuple[float, int]

class MFE_calculator.RNASolution(rna)

Bases: RNASolutionBase2

An RNA folding solution and working data

An instance holds the working data used when solving an RNA folding problem, and afterwards to generate. various expressions of the solution.

Parameters:

rna (str)

as_dots()

Dot-bracket form of the fold

walk_range(i, j)

Walk the range [i,j] inclusive using min(U,V).

We use this when backtracking the energy calculation to discover the lowest energy structure, in contexts where we know the subsequent energy calculation will have chosen the lesser of the closed (j pairs with i) and open (j pairs with something else or is unpaired) energy.

walk_range_ex(i, j, closed)

Walk the range [i,j] inclusive specifying open or closed.

We use this when backtracking the energy calculation to discover the lowest energy structure. Nodes (the data for a range) in our algorithm offer both closed (j pairs with a) and open (j pairs with something else or is unpaired) energy, and a subsequent nodes choose between them in offering their own energies. When backtracking, determining the structure of [i,j] in the minimum energy configuration may be informed by knowledge of that choice, provided through the ‘closed’ argument.

Parameters:

  • i (int)

  • j (int)

class MFE_calculator.RNASolutionBase(rna)

An RNA folding solution and working data

An instance holds the working data used when solving an RNA folding problem, and afterwards to generate. various expressions of the solution.

Parameters:

rna (str)

K
M
U
V
energy()

The free energy of the minimum energy solution

getK(i, j)
Return type:

int

getM(i, j)
Return type:

int

getU(i, j)
Return type:

float

getV(i, j)
Return type:

float

irna
class MFE_calculator.RNASolutionBase2(rna)

Bases: RNASolutionBase

An RNA folding solution and working data

An instance holds the working data used when solving an RNA folding problem, and afterwards to generate. various expressions of the solution.

Parameters:

rna (str)

branches(i, j)

Enumerate the branches in the open interval [i,j].

When we consider the energy freed by potential pair (i-1,j+1), we must understand what open structure it “closes”. It is enough to list the branches (helices) chosen when forming the open conformation of [i,j], each identified by its initial pair, in ascending order of first member.

The number of branches distingishes 0: hairpin, 1: internal or bulge loop, and >1: multiloop, while in case 1 the differences between the ends of the proposed closing and those of the single base pair distinguish amongst bulges, interior and muylti-branch loops. A single branch does not signify helix continuation, since that is a closed structure.

Parameters:

  • i (int)

  • j (int)

Return type:

list[tuple]

get_arrays()

Get references to the K, U and V arrays

Return type:

tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray]

MFE_calculator._PAIRS = ('AU', 'CG', 'GC', 'GU', 'UA', 'UG')