Data object for containing sets of multiple sequence alignments.

Allows for integration of sets of multiple sequence alignments

AAtoDNA(f_AA_aligned, f_DNA_unaligned, f_DNA_out)¶

Create a DNA multiple sequence alignment

from an amino acid multiple sequence alignment

Parameters:	f_AA_aligned – File containing aligned amino acid sequences f_DNA_unaligned – The file containing unaligned DNA sequences f_DNA_out – The desired output DNA filename.

Note

This function requires pal2nal.

align(filename1, filename2=None, AAtoDNA=True)¶

Align orthologous sequences.

Parameters:	filename1 – Output filename for primary alignments filename2 – Output filename for secondary alignments AAtoDNA – Specifies that secondary sequences are DNA and should be aligned based on AA alignment.

alignfunc(f_in, f_out, c=5, ir=500, **kwargs)¶

Create multiple sequence alignment from unaligned sequences

Parameters:	f_in – The file of unaligned sequence. f_out – The desired output filename. ir – Specifies the -ir flag to msaprobs c – Specifies the -c flag to msaprobs

Note

This function requires msaprobs.

calc_sim_mat_pairwise()¶

Internal function: For pairwise optimization,

calculate the similarity matrix that will define the cluster of sequences.

Parameters:	self.multiMSA.MSAdict1 – A dictionary of dictionaries of sequences. self.multiMSA.methodnames1 – A list of the names of the methods producing each MSA. self.edgefunc – The function used to calculate the similarity between two sequences.
Returns:	A (nmethods*nspec) x (nmethods*nspec) matrix of edgeweights, stored to self.edgeweights_pairwise

Note

Sequences are blocked by method (according to self.methodnames).

These blocks are ordered by the specified species order (self.allspecs).

calc_sim_mat_toref()¶

Internal function: For “to reference” optimization,

calculate the similarity vector that will relate each sequence to the reference.

Parameters:	self.multiMSA.MSAdict1 – A dictionary of dictionaries of sequences. self.multiMSA.methodnames1 – A list of the names of the methods producing each MSA. self.edgefunc – The function used to calculate the similarity between two sequences.
Returns:	A (nspec) x (nspec) matrix of edgeweights, stored to self.edgeweights_toref

Note

This is the stage at which filtering takes place. Any sequence below the similarity cutoff is not assigned to the self.edgeweights_toref matrix.

getbitscore(seq1, seq2)¶

Calculate a bitscore between two (unaligned) sequences.

Parameters:	seq1 – Sequence 1 seq2 – Sequence 2
Variables:	self.AA – True if sequence alphabet is amino acids, false otherwise. self.stretcher_gapopen – The penalty for opening a gap in the alignment. self.stretcher_gapextend – The penalty for extending a gap in the alignment. self.scoremat – The score matrix to use for the scoring of the alignment.
Returns:	A bit score for the aligned sequences.

Note

pandas is required to manage scoring matrices.

getperID(seq1, seq2)¶

Calculate percent identity between two (unaligned) sequences.

Parameters:	seq1 – Sequence 1 seq2 – Sequence 2
Variables:	self.AA – True if sequence alphabet is amino acids, false otherwise. self.ignoregaps – True if gaps in the first (reference) sequence are to be ignored.
Returns:	A percent identity for the aligned sequences.

opt_cluster_toref()¶

Internal function: Takes sequences from each species with

the highest similarity to the reference.

optimize_cluster()¶

Internal function: optimize the sequence cluster.

If self.optrule is ‘pairwise’, optimize cluster by picking the

sequence for each species that optimizes the pairwise distance to current best sequences. This is repeated cyclically until convergence is reached.

If self.optrule is ‘toref’, pick the sequence from each species

that is most similar to the reference sequence

If self.optrule is ‘custom’, apply the function defined

in self.customoptfunc to the self.edgeweights_toref and/or self.edgeweights_pairwise matrices

optloop_pairwise()¶

Internal function: optimize cluster using pairwise similarities

and Gibbs sampling.

write_unaligned(filename1, filename2=None, inclspec=False, inclmet=False, specorder=None, labelfunc=None)¶

Write (unaligned) optimal sequences to a file.

Parameters:

filename1 – The file to which to write the primary MSAlist filename2 – The file to which to write the secondary MSAlist inclspec – Whether to include the species name in the sequence labels inclmet – Whether to include the method name in the sequence labels specorder – If specified, a different order for species in the output labelfunc – If specified, a function to output sequence labels. Should be of the form labelfunc(seq.name, species, method)