comparison barcode_splitter_multi/barcode_splitter.py @ 1:e3f91eee8c75 draft

Uploaded

0:50df2d629d51

import re
import sys
from collections import defaultdict
import subprocess

__version__      = "0.12"
__author__       = "Lance Parsons & Robert Leach"
__author_email__ = "[e-mail],[e-mail]"
__copyright__    = "Copyright 2011, Lance Parsons & Robert leach"
__license__      = ("BSD 2-Clause License "
                    "http://www.opensource.org/licenses/BSD-2-Clause")

        if(barcode_path is None):
            cur_outputs = unmatchedOutputs
            unmatched_path = getBarcodeMatchPath(approx_bc_dict,index_seqs)
	    incrementNDDictInt(unmatched_counts,unmatched_path)
        elif(MULTIMATCHED in barcode_path):
            cur_outputs = multimatchedOutputs
            unmatched_path = getBarcodeMatchPath(approx_bc_dict,barcode_path)
	    incrementNDDictInt(unmatched_counts,unmatched_path)
            sys.stderr.write('WARNING: More than one barcode matches %s, '

    "UNMATCHED", and "MULTIMATCHED" values) to use in the summary output.  The
    intent is to give the user useful feedback about which level(s) of barcodes
    are not matrching instead of listing all sequences as simply "UNMATCHED".
    It also reduces complexity of the unmatched output by not including
    specific barcodes, which do not matter in unmatched/multimatched cases.'''
    cur_dict = in_dict
    dim = 0
    path = []
    for cur_key in keys_list:
        dim += 1
        if cur_key is MATCHED or cur_key is MULTIMATCHED:
            path.append(cur_key)
        elif cur_key in cur_dict:
            path.append(MATCHED)
        else:
            path.append(UNMATCHED)
        cur_dict = in_dict[ANY][str(dim)]
    return path

def setNDApproxDictVal(in_dict, keys_list, mismatches):
    '''Creates (or builds upon) an N-Dimensional ("ND") dictionary (in_dict)
    whose keys are mismatch variants (constructed using the keys_list and the

1:e3f91eee8c75

import re
import sys
from collections import defaultdict
import subprocess

__version__      = "0.13"
__author__       = "Lance Parsons & Robert Leach"
__author_email__ = "[e-mail],[e-mail]"
__copyright__    = "Copyright 2011, Lance Parsons & Robert leach"
__license__      = ("BSD 2-Clause License "
                    "http://www.opensource.org/licenses/BSD-2-Clause")

        if(barcode_path is None):
            cur_outputs = unmatchedOutputs
            unmatched_path = getBarcodeMatchPath(approx_bc_dict,index_seqs)
	    incrementNDDictInt(unmatched_counts,unmatched_path)
            if (not UNMATCHED in unmatched_path and
                not MULTIMATCHED in unmatched_path):
                sys.stderr.write('WARNING: Sequences match barcodes on '
                                 'different rows: %s for sequence ID: %s\n'
                                 %(index_seqs, prim_index_read['seq_id']))
        elif(MULTIMATCHED in barcode_path):
            cur_outputs = multimatchedOutputs
            unmatched_path = getBarcodeMatchPath(approx_bc_dict,barcode_path)
	    incrementNDDictInt(unmatched_counts,unmatched_path)
            sys.stderr.write('WARNING: More than one barcode matches %s, '

    "UNMATCHED", and "MULTIMATCHED" values) to use in the summary output.  The
    intent is to give the user useful feedback about which level(s) of barcodes
    are not matrching instead of listing all sequences as simply "UNMATCHED".
    It also reduces complexity of the unmatched output by not including
    specific barcodes, which do not matter in unmatched/multimatched cases.'''
    dim = 0
    path = []
    for cur_key in keys_list:
        dim += 1
        if cur_key is MATCHED or cur_key is MULTIMATCHED:
            path.append(cur_key)
        elif cur_key in in_dict[ANY][str(dim)]:
            path.append(MATCHED)
        else:
            path.append(UNMATCHED)
    return path

def setNDApproxDictVal(in_dict, keys_list, mismatches):
    '''Creates (or builds upon) an N-Dimensional ("ND") dictionary (in_dict)
    whose keys are mismatch variants (constructed using the keys_list and the