BioPythonTip for GenBank

파싱한 후 취할 수 있는 객체들

RecordParser를 썼을때

parser = GenBank.RecordParser()
iterator = GenBank.Iterator()
while 1:
    cur_record = iterator.next()
    if cur_record is None:
        break
    cur_record.어쩌구저쩌구

위 파서는 핵산DB, 단백질 DB모두에게 적용된다.

• locus - The name specified after the LOCUS keyword in the GenBank record. This may be the accession number, or a clone id or something else.

• size - The size of the record.
• residue_type - The type of residues making up the sequence in this record. Normally something like RNA, DNA or PROTEIN, but may be as esoteric as 'ss-RNA circular'.

• data_file_division - The division this record is stored under in GenBank (ie. PLN -> plants; PRI -> humans, primates; BCT -> bacteria...)

• date - The date of submission of the record, in a form like '28-JUL-1998'
• accession - list of all accession numbers for the sequence.
• nid - Nucleotide identifier number.
• pid - Proteint identifier number
• version - The accession number + version (ie. AB01234.2)
• db_source - Information about the database the record came from
• gi - The NCBI gi identifier for the record.
• keywords - A list of keywords related to the record.
• segment - If the record is one of a series, this is info about which segment this record is (something like '1 of 6').
• source - The source of material where the sequence came from.
• organism - The genus and species of the organism (ie. 'Homo sapiens')
• taxonomy - A listing of the taxonomic classification of the organism, starting general and getting more specific.
• references - A list of Reference objects.
  • number - The number of the reference in the listing of references.
  • bases - The bases in the sequence the reference refers to.
  • authors - String with all of the authors.
  • title - The title of the reference.
  • journal - Information about the journal where the reference appeared.
  • medline_id - The medline id for the reference.
  • pubmed_id - The pubmed_id for the reference.
  • remark - Free-form remarks about the reference.
• comment - Text with any kind of comment about the record.
• features - A listing of Features making up the feature table.
  • key - The key name of the featue (ie. source)
  • location - The string specifying the location of the feature.
  • qualfiers - A listing Qualifier objects in the feature.
    • key - The key name of the qualifier (ie. /organim=)
    • value - The value of the qualifier ("Dictyostelium discoideum").
• base_counts - A string with the counts of bases for the sequence.
• origin - A string specifying info about the origin of the sequence.
• sequence - A string with the sequence itself.

FeatureParser 를 썼을때

parser = GenBank.FeatureParser()
iterator = GenBank.Iterator()
while 1:
    cur_record = iterator.next()
    if cur_record is None:
        break
    cur_record.어쩌구저쩌구

가장 자세히 파싱할 수 있는 파서이며, 위 cur_record 객체는 SeqFeature를 따르는 객체로, BioCorba로 연동도 될 수 있다.

• features
  • location - the location of the feature on the sequence
    • position - The position of the boundary.
    • extension - An optional argument which must be zero since we don't have an extension. The argument is provided so that the same number of arguments can be passed to all position types.
  • type - the specified type of the feature (ie. CDS, exon, repeat...)
  • ref - A reference to another sequence. This could be an accession number for some different sequence.
  • ref_db - A different database for the reference accession number.

  • qualifier - A dictionary of qualifiers on the feature. These are analagous to the qualifiers from a GenBank feature table. The keys of the dictionary are qualifier names, the values are the qualifier values.

  • sub_features - Additional SeqFeatures which fall under this 'parent' feature. For instance, if we having something like:

    • CDS    join(1..10,30..40,50..60) The the top level feature would be a CDS from 1 to 60, and the sub features would be of 'CDS_span' type and would be from 1 to 10, 30 to 40 and 50 to 60, respectively.

• references
  • location - A list of Location objects specifying regions of the sequence that the references correspond to. If no locations are specified, the entire sequence is assumed.
  • authors - A big old string, or a list split by author, of authors for the reference.
  • title - The title of the reference.
  • journal - Journal the reference was published in.
  • medline_id - A medline reference for the article.
  • pubmed_id - A pubmed reference for the article.
  • comment - A place to stick any comments about the reference.
• name : Version
• annotations
• discription : Definition