
:>"^Å  ã               @   sK  d  Z  d d l m Z d d l m Z e j e j j ƒ  k sD t ‚ Gd d „  d e j ƒ Z	 e	 ƒ  Z
 e j e j j ƒ  k s„ t ‚ Gd d „  d e	 ƒ Z e ƒ  Z Gd d	 „  d	 e j ƒ Z e ƒ  Z Gd
 d „  d e ƒ Z e ƒ  Z Gd d „  d e j ƒ Z e ƒ  Z Gd d „  d e j ƒ Z e ƒ  Z Gd d „  d e ƒ Z e ƒ  Z d S)z;Standard nucleotide and protein alphabets defined by IUPAC.é    )ÚAlphabet)Ú	IUPACDatac               @   s   e  Z d  Z d Z e j Z d S)ÚExtendedIUPACProteinal  Extended uppercase IUPAC protein single letter alphabet including X etc.

    In addition to the standard 20 single letter protein codes, this includes:

    - ``B`` = "Asx";  Aspartic acid (R) or Asparagine (N)
    - ``X`` = "Xxx";  Unknown or 'other' amino acid
    - ``Z`` = "Glx";  Glutamic acid (E) or Glutamine (Q)
    - ``J`` = "Xle";  Leucine (L) or Isoleucine (I), used in mass-spec (NMR)
    - ``U`` = "Sec";  Selenocysteine
    - ``O`` = "Pyl";  Pyrrolysine

    This alphabet is not intended to be used with ``X`` for Selenocysteine
    (an ad-hoc standard prior to the IUPAC adoption of ``U`` instead).
    N)Ú__name__Ú
__module__Ú__qualname__Ú__doc__r   Úextended_protein_lettersÚletters© r   r   ú7/tmp/pip-build-ww9dw3qa/biopython/Bio/Alphabet/IUPAC.pyr      s   r   c               @   s   e  Z d  Z d Z e j Z d S)ÚIUPACProteinz]IUPAC protein alphabet of the 20 standard amino acids.

    Uppercase and single letter.
    N)r   r   r   r   r   Úprotein_lettersr
   r   r   r   r   r   /   s   r   c               @   s   e  Z d  Z d Z e j Z d S)ÚIUPACAmbiguousDNAzUppercase IUPAC ambiguous DNA.N)r   r   r   r   r   Zambiguous_dna_lettersr
   r   r   r   r   r   ?   s   r   c               @   s   e  Z d  Z d Z e j Z d S)ÚIUPACUnambiguousDNAz4Uppercase IUPAC unambiguous DNA (letters GATC only).N)r   r   r   r   r   Zunambiguous_dna_lettersr
   r   r   r   r   r   H   s   r   c               @   s   e  Z d  Z d Z e j Z d S)ÚExtendedIUPACDNAzÔExtended IUPAC DNA alphabet.

    In addition to the standard letter codes GATC, this includes:

    - ``B`` = 5-bromouridine
    - ``D`` = 5,6-dihydrouridine
    - ``S`` = thiouridine
    - ``W`` = wyosine

    N)r   r   r   r   r   Zextended_dna_lettersr
   r   r   r   r   r   R   s   
r   c               @   s   e  Z d  Z d Z e j Z d S)ÚIUPACAmbiguousRNAzUppercase IUPAC ambiguous RNA.N)r   r   r   r   r   Zambiguous_rna_lettersr
   r   r   r   r   r   f   s   r   c               @   s   e  Z d  Z d Z e j Z d S)ÚIUPACUnambiguousRNAz4Uppercase IUPAC unambiguous RNA (letters GAUC only).N)r   r   r   r   r   Zunambiguous_rna_lettersr
   r   r   r   r   r   o   s   r   N)r   ZBior   ZBio.Datar   r	   ÚupperÚAssertionErrorZProteinAlphabetr   Zextended_proteinr   r   ZproteinZDNAAlphabetr   Zambiguous_dnar   Zunambiguous_dnar   Zextended_dnaZRNAAlphabetr   Zambiguous_rnar   Zunambiguous_rnar   r   r   r   Ú<module>   s&   							