o RŀgJG@sdZddlZddlmZddlmZddlmZddlmZddl m Z ddl m Z dd l m Z dd l mZdd lmZGd d d eZdddZddZddZddZddZddZddZedkrpddlmZedSdS) zCode for dealing with Codon Alignment. CodonAlignment class is inherited from MultipleSeqAlignment class. This is the core class to deal with codon alignment in biopython. N)erfc)sqrt)BiopythonWarning)MultipleSeqAlignment)_get_codon_list) cal_dn_ds)CodonSeq) CodonTable) SeqRecordc@sdeZdZdZdddZddZdd Zd d Zd d ZddZ dddZ dddZ e ddZ dS)CodonAlignmentaCodon Alignment class that inherits from MultipleSeqAlignment. >>> from Bio.SeqRecord import SeqRecord >>> a = SeqRecord(CodonSeq("AAAACGTCG"), id="Alpha") >>> b = SeqRecord(CodonSeq("AAA---TCG"), id="Beta") >>> c = SeqRecord(CodonSeq("AAAAGGTGG"), id="Gamma") >>> print(CodonAlignment([a, b, c])) CodonAlignment with 3 rows and 9 columns (3 codons) AAAACGTCG Alpha AAA---TCG Beta AAAAGGTGG Gamma NcCsFt|||D] }t|jtstdq|ddkr!tddS)zInitialize the class.zACodonSeq objects are expected in each SeqRecord in CodonAlignmentrzKAlignment length is not a multiple of three (i.e. a whole number of codons)N)r__init__ isinstanceseqr TypeErrorget_alignment_length ValueError)selfrecordsnamerecrQ/var/www/html/myenv/lib/python3.10/site-packages/Bio/codonalign/codonalignment.pyr)s  zCodonAlignment.__init__cstj}d|fg}|dkr#|fddjDn#|fddjddD|d|jjd dd d |S) aDReturn a multi-line string summary of the alignment. This output is indicated to be readable, but large alignment is shown truncated. A maximum of 20 rows (sequences) and 60 columns (20 codons) are shown, with the record identifiers. This should fit nicely on a single screen. e.g. z6CodonAlignment with %i rows and %i columns (%i codons)<cg|] }j|ddqSrlength _str_line.0rrrr Jz*CodonAlignment.__str__..crrrr!r#rrr$Lr%Nz...r )len_recordsrget_aln_lengthextendappendr join)rrowslinesrr#r__str__:s "  zCodonAlignment.__str__cst|tr |j|St|trt|j|St|dkr td|\}t|tr0|j|SttrDdfdd|j|DStfdd|j|DS)z3Return a CodonAlignment object for single indexing.zInvalid index type.r c3s|] }t|VqdSN)strr! col_indexrr ^z-CodonAlignment.__getitem__..c3s|]}|VqdSr3rr!r5rrr7`s ) rintr*slicer r)rr.r)rindex row_indexrr5r __getitem__Qs       zCodonAlignment.__getitem__cCst|tr%t|t|krtdtdtddt||D}t|St|tr CodonAlignment * CodonAlignment + MultipleSeqAlignment -> MultipleSeqAlignment zTWhen adding two alignments they must have the same length (i.e. same number or rows)zsPlease make sure the two CodonAlignment objects are sharing the same codon table. This is not checked by Biopython.css*|]\}}tt|j|jdVqdS))rN)r rr)r"leftrightrrrr7ws  z)CodonAlignment.__add__..z^Only CodonAlignment or MultipleSeqAlignment object can be added with a CodonAlignment object. z detected.) rr r)rwarningswarnrziprtoMultipleSeqAlignmentrobject)rothermergedrrr__add__ds0   zCodonAlignment.__add__cCs |dS)zGet alignment length.r rr#rrrr+s zCodonAlignment.get_aln_lengthcCsdd|jD}t|S)zConvert the CodonAlignment to a MultipleSeqAlignment. Return a MultipleSeqAlignment containing all the SeqRecord in the CodonAlignment using Seq to store sequences cSs g|] }t|j|jdqSid)r rtoSeqrKr!rrrr$s z9CodonAlignment.toMultipleSeqAlignment..)r*r)r alignmentsrrrrCsz%CodonAlignment.toMultipleSeqAlignmentNG86cCsddlm}|durtjd}dd|jD}t|j}g}g}t|D]F}|g|gt|dD]3} || kr[t|j||j| ||d\} } ||| ||| q6||d||dq6q$|||d } |||d } | | fS) zAvailable methods include NG86, LWL85, YN00 and ML. Argument: - method - Available methods include NG86, LWL85, YN00 and ML. - codon_table - Codon table to use for forward translation. r)DistanceMatrixNcSsg|]}|jqSrrJr"irrrr$sz3CodonAlignment.get_dn_ds_matrix..method codon_tableg)matrix) Bio.Phylo.TreeConstructionrOr generic_by_idr*r)ranger-r)rrTrUDMnamessize dn_matrix ds_matrixrRjdndsdn_dmds_dmrrrget_dn_ds_matrixs4        zCodonAlignment.get_dn_ds_matrixUPGMAc Csddlm}|durtjd}|j||d\}}|}|}|dkr0||} ||} | | fS|dkrB||} ||} | | fStd|d ) zConstruct dn tree and ds tree. Argument: - dn_ds_method - Available methods include NG86, LWL85, YN00 and ML. - tree_method - Available methods include UPGMA and NJ. r)DistanceTreeConstructorNrPrSreNJzUnknown tree method (z"). Only NJ and UPGMA are accepted.)rWrfr rXrdupgmanj RuntimeError) r dn_ds_method tree_methodrUrfrbrcdn_constructords_constructordn_treeds_treerrrget_dn_ds_trees&         zCodonAlignment.get_dn_ds_treecCsdd|jD}||S)zConvert a MultipleSeqAlignment to CodonAlignment. Function to convert a MultipleSeqAlignment to CodonAlignment. It is the user's responsibility to ensure all the requirement needed by CodonAlignment is met. cSs$g|]}ttt|j|jdqSrI)r rr4rrKrQrrrr$s$z+CodonAlignment.from_msa..)r*)clsalignrrrrfrom_msaszCodonAlignment.from_msa)r N)rNN)rNreN)__name__ __module__ __qualname____doc__rr1r=rGr+rCrdrq classmethodrtrrrrr s % %  r 皙?csddl}|dur tjd}tdd|Dstddd|D}tt|dkr-td |dd }||j }|j D]}d ||<q.z#mktest accepts CodonAlignment list.cSsg|]}|qSrrHrQrrrr$zmktest..z;CodonAlignment object for mktest should be of equal length.r stopr'csh|]}|qSrrr"krRrr r{zmktest..)rrrr)rU-css|] }t|dkVqdS)rPN)r)r}rrrr7r8)copyr rXallrr)setrjdeepcopy forward_table stop_codonsr-rrrY_get_codon2codon_matrixunion _get_subgraph_count_replacement_G_test) codon_alnsrUalphar codon_aln_len codon_num codon_dictr| codon_lst codon_aln codon_set uniq_codonsr_ uniq_codonsyn_fix nonsyn_fixsyn_poly nonsyn_polyGnonsyn_G all_codon fix_or_notnonsyn_subgraphsubgraphthis_nonthis_synrrrmktests`                   rcCsddl}d}ddt|j|jD}||j}|jD]}d||<qt|}i}i}i} i} t|D]O\} } i| | <i| | <t| D]>\} }|D]7}| d| || | dd}|| ||krrd| | |<d| | |<qK| |krd| | |<d| | |<qKqEq5|D]4}i||<i||<|D]'}||krd|||<d|||<qt| |||||<t| |||||<qq||fS) zGet codon codon substitution matrix (PRIVATE). Elements in the matrix are number of synonymous and nonsynonymous substitutions required for the substitution. rN)ATCrcSsg|]}d|vr|qS)UrrQrrrr$4s z+_get_codon2codon_matrix..r|rPg?) rlistrkeysrrr) enumerate _dijkstra)rUr base_tuplecodonsrr|numrrgraph graph_nonsynrRcodonpbr_ tmp_codoncodon1codon2rrrr+sN         rcCsli}i}|D] }d||<d||<qd||<t|}t|dkrtd}d}|D]}|dur6||}|}q)|||krB||}|}q)||||D]\} } || ||| krh||| || <||| <qN||krnnt|dks#g} |}d} ||kr| |dkr| d|||}nn||ks~| d|tt| dD]} | || | | | d7} q| S)aDijkstra's algorithm Python implementation (PRIVATE). Algorithm adapted from http://thomas.pelletier.im/2010/02/dijkstras-algorithm-python-implementation/. However, an obvious bug in:: if D[child_node] >(<) D[node] + child_value: is fixed. This function will return the distance between start and end. Arguments: - graph: Dictionary of dictionary (keys are vertices). - start: Start vertex. - end: End vertex. Output: List of vertices from the beginning to the end. dr rNrP)rrr)removeitemscountinsertrY)rstartendDPnode unseen_nodesshortest temp_node child_node child_valuepathdistancerRrrrr]sP          rcCsXddlm}t|dkrdSt|dkr$t|}|||d|dSt|}t|S)z9Count replacement needed for a given codon_set (PRIVATE).rfloorrP)rrr2)mathrr)r_prim)rrrrrrrrs   rcCsddlm}ddlm}ddlm}ddlm}ddlm}g}g}|D]0}| |||D]$} || ||| f|vrU| |||| f|vrU| || ||| fq1q&|t } |D]\} } } | |  | | | f| |  | | | fq]g}t |d}| |ddd}|||r||\}} } | |vr| | | | | |f| | D] }|d|vr|||q|sd}|D] }|||d7}q|S) zPrim's algorithm to find minimum spanning tree (PRIVATE). Code is adapted from http://programmingpraxis.com/2010/04/09/minimum-spanning-tree-prims-algorithm/ r) defaultdict)heapify)heappop)heappushrNr2) collectionsrheapqrrrrrrr-rradd)rrrrrrnodesedgesrRr_connn1n2cmstused usable_edgescosterrrrrrsH        ,     rcCs@i}|D]}i||<|D]}||kr||||||<q q|S)z>> print("%0.6f" % _G_test([17, 7, 42, 2])) 0.004924 r)logr2rPr N)rrsumrBrr) site_countsrrtottot_syntot_nontot_fixtot_polyexpobsexrrrrs      r__main__) run_doctest)Nrz)rxr@rrrBior Bio.AlignrBio.codonalign.codonseqrrrBio.Datar Bio.SeqRecordr r rrrrrrrru Bio._utilsrrrrrs0          ND2D)