bam.py 文件源码

def create_bam_outfile(file_name, chrom_names, chrom_lengths, template=None, pgs=None, cos=None, rgs=None, replace_rg=False):
    """ Creates a bam file with given chromosome names and lengths.
    template is an existing bam file.  If it is specified, chrom_names and chrom_lengths
    are ignored. pg is dictionary specifying a 'PG' entry. ID field is required; PN/CL/PP/DS/VN fields are optional.
    rgs is a list of dicts specifiying an 'RG' entry. If replace_rg is True, the existing 'RG' entry is overwritten.
    """
    if template:
        header = template.header
        if pgs is not None:
            for pg in pgs:
                if not header.has_key('PG'):
                    header['PG'] = []
                # add in the PP field based on previous PG entry
                if len(header['PG']) > 0:
                    pp = header['PG'][-1]['ID']
                    if pp is not None:
                        pg['PP'] = pp
                header['PG'].append(pg)

        if cos is not None:
            for co in cos:
                if not header.has_key('CO'):
                    header['CO'] = []
                header['CO'].append(co)

        if rgs is not None:
            if replace_rg and header.has_key('RG') and len(rgs) > 0:
                header['RG'] = []
            for rg in rgs:
                if not header.has_key('RG'):
                    header['RG'] = []
                header['RG'].append(rg)

        bam_file = pysam.Samfile(file_name, 'wb', header=header)
        tids = {name:n for (n, name) in enumerate(template.references)}
    else:
        header = {'SQ': [{'SN': chrom_names[n], 'LN': chrom_lengths[n]} for n in xrange(len(chrom_names))]}
        bam_file = pysam.Samfile(file_name, 'wb', header=header)
        tids = {chrom_names[n]:n for n in xrange(len(chrom_names))}
    return bam_file, tids