Alignment: API rework. 'obi align' is now 'obi lcs', and the results are
now written to columns automatically created in the output view, all optimally handled at the C level.
This commit is contained in:
Celine Mercier
@ -1,120 +0,0 @@
|
||||
from obitools3.apps.progress cimport ProgressBar # @UnresolvedImport
|
||||
from obitools3.obidms._obidms import OBIDMS, OBIView # TODO cimport doesn't work
|
||||
|
||||
|
||||
import time
|
||||
|
||||
__title__="Aligns one sequence column with itself or two sequence columns"
|
||||
|
||||
|
||||
default_config = { 'inputview' : None,
|
||||
}
|
||||
|
||||
def addOptions(parser):
|
||||
|
||||
# TODO put this common group somewhere else but I don't know where
|
||||
group=parser.add_argument_group('DMS and view options')
|
||||
|
||||
group.add_argument('--default-dms','-d',
|
||||
action="store", dest="obi:defaultdms",
|
||||
metavar='<DMS NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the default DMS for reading and writing data.")
|
||||
|
||||
group.add_argument('--input-view','-i',
|
||||
action="store", dest="obi:inputview",
|
||||
metavar='<INPUT VIEW NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the input view.")
|
||||
|
||||
# TODO eventually 2nd view, or 2nd column?
|
||||
|
||||
group.add_argument('--output-view','-o',
|
||||
action="store", dest="obi:outputview",
|
||||
metavar='<OUTPUT VIEW NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the output view.")
|
||||
|
||||
|
||||
group=parser.add_argument_group('obi align specific options')
|
||||
|
||||
group.add_argument('--lcs','-C',
|
||||
action="store", dest="align:alitype",
|
||||
metavar='<ALIGNMENT TYPE>',
|
||||
default='lcs',
|
||||
type=str,
|
||||
help="Compute alignment using the LCS method (default).")
|
||||
|
||||
group.add_argument('--threshold','-t',
|
||||
action="store", dest="align:threshold",
|
||||
metavar='<THRESHOLD>',
|
||||
default=0.0,
|
||||
type=float,
|
||||
help="Score threshold. If the score is normalized and expressed in similarity (default),"
|
||||
" it is an identity, e.g. 0.95 for an identity of 95%%. If the score is normalized"
|
||||
" and expressed in distance, it is (1.0 - identity), e.g. 0.05 for an identity of 95%%."
|
||||
" If the score is not normalized and expressed in similarity, it is the length of the"
|
||||
" Longest Common Subsequence. If the score is not normalized and expressed in distance,"
|
||||
" it is (reference length - LCS length)."
|
||||
" Only sequence pairs with a similarity above <THRESHOLD> are printed. Default: 0.00"
|
||||
" (no threshold).")
|
||||
|
||||
group.add_argument('--longest_length','-L',
|
||||
action="store_const", dest="align:reflength",
|
||||
default=0,
|
||||
const=1,
|
||||
help="The reference length is the length of the longest sequence."
|
||||
" Default: the reference length is the length of the alignment.")
|
||||
|
||||
group.add_argument('--shortest_length','-l',
|
||||
action="store_const", dest="align:reflength",
|
||||
default=0,
|
||||
const=2,
|
||||
help="The reference length is the length of the shortest sequence."
|
||||
" Default: the reference length is the length of the alignment.")
|
||||
|
||||
group.add_argument('--raw','-r',
|
||||
action="store_false", dest="align:normalize",
|
||||
default=True,
|
||||
help="Raw score, not normalized. Default: score is normalized with the reference sequence length.")
|
||||
|
||||
group.add_argument('--distance','-D',
|
||||
action="store_false", dest="align:similarity",
|
||||
default=True,
|
||||
help="Score is expressed in distance. Default: score is expressed in similarity.")
|
||||
|
||||
|
||||
|
||||
def run(config):
|
||||
|
||||
# Open DMS
|
||||
d = OBIDMS(config['obi']['defaultdms'])
|
||||
|
||||
# Open input view 1
|
||||
iview = d.open_view(config['obi']['inputview'])
|
||||
|
||||
# TODO Open input view 2 if there is one
|
||||
|
||||
# Create output view
|
||||
oview = d.new_view(config['obi']['outputview'])
|
||||
|
||||
# TODO Take other alignment types into account when they'll be implemented
|
||||
|
||||
# Call cython alignment function
|
||||
iview.align(oview, threshold=config['align']['threshold'], normalize=config['align']['normalize'], reference=config['align']['reflength'], similarity_mode=config['align']['similarity'])
|
||||
|
||||
print(repr(oview))
|
||||
|
||||
iview.close()
|
||||
oview.close()
|
||||
d.close()
|
||||
|
||||
print("Done.")
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
209
python/obitools3/commands/lcs.pyx
Normal file
209
python/obitools3/commands/lcs.pyx
Normal file
@ -0,0 +1,209 @@
|
||||
#cython: language_level=3
|
||||
|
||||
from obitools3.apps.progress cimport ProgressBar # @UnresolvedImport
|
||||
from obitools3.obidms._obidms cimport OBIDMS # TODO cimport doesn't work
|
||||
from obitools3.utils cimport str2bytes
|
||||
|
||||
from obitools3.obidms.capi.obialign cimport obi_lcs_align_one_column
|
||||
|
||||
|
||||
import time
|
||||
|
||||
__title__="Aligns one sequence column with itself or two sequence columns"
|
||||
|
||||
|
||||
default_config = { 'inputview' : None,
|
||||
}
|
||||
|
||||
def addOptions(parser):
|
||||
|
||||
# TODO put this common group somewhere else but I don't know where.
|
||||
# Also some options should probably be in another group
|
||||
group=parser.add_argument_group('DMS and view options')
|
||||
|
||||
group.add_argument('--default-dms', '-d',
|
||||
action="store", dest="obi:defaultdms",
|
||||
metavar='<DMS NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the default DMS for reading and writing data.")
|
||||
|
||||
group.add_argument('--input-view-1', '-i',
|
||||
action="store", dest="obi:inputview1",
|
||||
metavar='<INPUT VIEW NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the (first) input view.")
|
||||
|
||||
group.add_argument('--input-view-2', '-I',
|
||||
action="store", dest="obi:inputview2",
|
||||
metavar='<INPUT VIEW NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="Eventually, the name of the second input view.")
|
||||
|
||||
group.add_argument('--input-column-1', '-c',
|
||||
action="store", dest="obi:inputcolumn1",
|
||||
metavar='<INPUT COLUMN NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="Name of the (first) input column. "
|
||||
" Default: the default nucleotide sequence column of the view if there is one.")
|
||||
|
||||
group.add_argument('--input-column-2', '-C',
|
||||
action="store", dest="obi:inputcolumn2",
|
||||
metavar='<INPUT COLUMN NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="Eventually, the name of the second input column.")
|
||||
|
||||
group.add_argument('--input-elt-1', '-e',
|
||||
action="store", dest="obi:inputelement1",
|
||||
metavar='<INPUT ELEMENT NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="If the first input column has multiple elements per line, name of the element referring to the sequence to align. "
|
||||
" Default: the first element of the line.")
|
||||
|
||||
group.add_argument('--input-elt-2', '-E',
|
||||
action="store", dest="obi:inputelement2",
|
||||
metavar='<INPUT ELEMENT NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="If the second input column has multiple elements per line, name of the element referring to the sequence to align. "
|
||||
" Default: the first element of the line.")
|
||||
|
||||
group.add_argument('--id-column-1', '-f',
|
||||
action="store", dest="obi:idcolumn1",
|
||||
metavar='<ID COLUMN NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="Name of the (first) column containing the identifiers of the sequences to align. "
|
||||
" Default: the default ID column of the view if there is one.")
|
||||
|
||||
group.add_argument('--id-column-2', '-F',
|
||||
action="store", dest="obi:idcolumn2",
|
||||
metavar='<ID COLUMN NAME>',
|
||||
default="",
|
||||
type=str,
|
||||
help="Eventually, the name of the second ID column.")
|
||||
|
||||
group.add_argument('--output-view', '-o',
|
||||
action="store", dest="obi:outputview",
|
||||
metavar='<OUTPUT VIEW NAME>',
|
||||
default=None,
|
||||
type=str,
|
||||
help="Name of the output view.")
|
||||
|
||||
|
||||
group=parser.add_argument_group('obi lcs specific options')
|
||||
|
||||
group.add_argument('--threshold','-t',
|
||||
action="store", dest="align:threshold",
|
||||
metavar='<THRESHOLD>',
|
||||
default=0.0,
|
||||
type=float,
|
||||
help="Score threshold. If the score is normalized and expressed in similarity (default),"
|
||||
" it is an identity, e.g. 0.95 for an identity of 95%%. If the score is normalized"
|
||||
" and expressed in distance, it is (1.0 - identity), e.g. 0.05 for an identity of 95%%."
|
||||
" If the score is not normalized and expressed in similarity, it is the length of the"
|
||||
" Longest Common Subsequence. If the score is not normalized and expressed in distance,"
|
||||
" it is (reference length - LCS length)."
|
||||
" Only sequence pairs with a similarity above <THRESHOLD> are printed. Default: 0.00"
|
||||
" (no threshold).")
|
||||
|
||||
group.add_argument('--longest-length','-L',
|
||||
action="store_const", dest="align:reflength",
|
||||
default=0,
|
||||
const=1,
|
||||
help="The reference length is the length of the longest sequence."
|
||||
" Default: the reference length is the length of the alignment.")
|
||||
|
||||
group.add_argument('--shortest-length','-l',
|
||||
action="store_const", dest="align:reflength",
|
||||
default=0,
|
||||
const=2,
|
||||
help="The reference length is the length of the shortest sequence."
|
||||
" Default: the reference length is the length of the alignment.")
|
||||
|
||||
group.add_argument('--raw','-r',
|
||||
action="store_false", dest="align:normalize",
|
||||
default=True,
|
||||
help="Raw score, not normalized. Default: score is normalized with the reference sequence length.")
|
||||
|
||||
group.add_argument('--distance','-D',
|
||||
action="store_false", dest="align:similarity",
|
||||
default=True,
|
||||
help="Score is expressed in distance. Default: score is expressed in similarity.")
|
||||
|
||||
group.add_argument('--print-seq','-s',
|
||||
action="store_true", dest="align:printseq",
|
||||
default=False,
|
||||
help="The nucleotide sequences are written in the output view. Default: they are not written.")
|
||||
|
||||
group.add_argument('--print-count','-n',
|
||||
action="store_true", dest="align:printcount",
|
||||
default=False,
|
||||
help="Sequence counts are written in the output view. Default: they are not written.")
|
||||
|
||||
|
||||
cpdef align(str dms_n,
|
||||
str input_view_1_n, str output_view_n,
|
||||
str input_view_2_n="",
|
||||
str input_column_1_n="", str input_column_2_n="",
|
||||
str input_elt_1_n="", str input_elt_2_n="",
|
||||
str id_column_1_n="", str id_column_2_n="",
|
||||
double threshold=0.0, bint normalize=True,
|
||||
int reference=0, bint similarity_mode=True,
|
||||
bint print_seq=False, bint print_count=False,
|
||||
comments="") :
|
||||
|
||||
cdef OBIDMS d
|
||||
d = OBIDMS(dms_n)
|
||||
|
||||
# Align 1 column (2 columns not implemented yet)
|
||||
if obi_lcs_align_one_column(d._pointer, \
|
||||
str2bytes(input_view_1_n), \
|
||||
str2bytes(input_column_1_n), \
|
||||
str2bytes(input_elt_1_n), \
|
||||
str2bytes(id_column_1_n), \
|
||||
str2bytes(output_view_n), \
|
||||
str2bytes(comments), \
|
||||
print_seq, \
|
||||
print_count, \
|
||||
threshold, normalize, reference, similarity_mode) < 0 :
|
||||
raise Exception("Error aligning sequences")
|
||||
|
||||
d.close()
|
||||
|
||||
|
||||
def run(config):
|
||||
|
||||
# TODO: Build formatted comments with all parameters etc
|
||||
comments = "Obi align"
|
||||
|
||||
# Call cython alignment function
|
||||
align(config['obi']['defaultdms'], \
|
||||
config['obi']['inputview1'], \
|
||||
config['obi']['outputview'], \
|
||||
input_view_2_n = config['obi']['inputview2'], \
|
||||
input_column_1_n = config['obi']['inputcolumn1'], \
|
||||
input_column_2_n = config['obi']['inputcolumn2'], \
|
||||
input_elt_1_n = config['obi']['inputelement1'], \
|
||||
input_elt_2_n = config['obi']['inputelement2'], \
|
||||
id_column_1_n = config['obi']['idcolumn1'], \
|
||||
id_column_2_n = config['obi']['idcolumn2'], \
|
||||
threshold = config['align']['threshold'], \
|
||||
normalize = config['align']['normalize'], \
|
||||
reference = config['align']['reflength'], \
|
||||
similarity_mode = config['align']['similarity'], \
|
||||
print_seq = config['align']['printseq'], \
|
||||
print_count = config['align']['printcount'], \
|
||||
comments = comments)
|
||||
|
||||
print("Done.")
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user