obitools/obitools3
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 66 Wiki Activity

1st version of obi align command and reworked functions that handle

Browse Source 
column alignment
This commit is contained in:
Celine Mercier
2016-08-10 14:51:02 +02:00
parent 26b8e1f215
commit 2aaa87edcc
8 changed files with 373 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

128
python/obitools3/commands/align.pyx Normal file
View File

@ -0,0 +1,128 @@
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,
'skip' : 0,
'only' : None,
'skiperror' : False,
'moltype' : 'nuc',
}
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, either raw if the view is in the default DMS,"
" or in the form 'dms:view' if it is in another DMS.")
# 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, either raw if the view is in the default DMS,"
" or in the form 'dms:view' if it is in another DMS.")
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.")
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="ali",
const="longest",
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="ali",
const="shortest",
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):
#pb = ProgressBar(1, config, seconde=5) # TODO
# 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 if necessary
if config['obi']['outputview'] is not None :
oview = d.new_view(config['obi']['outputview'])
else :
oview = None
# TODO Take other alignment types into account when they'll be implemented
# Call cython alignment function
iview.align(output_view=oview)
print("Done.")

10
python/obitools3/obidms/_obidms.pxd
View File

@ -77,7 +77,15 @@ cdef class OBIView_NUC_SEQS(OBIView):
cdef OBIDMS_column qualities
cpdef delete_column(self, str column_name)
cpdef align(self,
OBIView iview2=*,
object output_view=*,
double threshold=*,
bint normalize=*,
int reference=*,
bint similarity_mode=*
)
cdef class OBIView_line :

76
python/obitools3/obidms/_obidms.pyx
View File

@ -10,7 +10,9 @@ from .capi.obidmscolumn cimport obi_close_column, \
OBIDMS_column_header_p
from .capi.obiutils cimport obi_format_date
from .capi.obialign cimport obi_align_one_column
from .capi.obitypes cimport const_char_p, \
OBIType_t, \
OBI_INT, \
@ -454,6 +456,7 @@ cdef class OBIView :
for line in self.__iter__() :
to_print = to_print + str(line) + "\n"
return to_print
#############################################
@ -524,7 +527,76 @@ cdef class OBIView_NUC_SEQS(OBIView):
def __setitem__(self, index_t line_idx, OBI_Nuc_Seq sequence_obj) :
for key in sequence_obj :
self[line_idx][key] = sequence_obj[key]
# TODO
cpdef align(self, OBIView iview2=None, object output_view=None,
double threshold=0.0, bint normalize=True, int reference=0, bint similarity_mode=True) :
cdef OBIView iview1
cdef OBIView oview
cdef Obiview_p iview1_p
cdef Obiview_p iview2_p
cdef Obiview_p oview_p
cdef OBIDMS_column icol1
cdef OBIDMS_column_p icol1_p
cdef OBIDMS_column_p* icol1_pp
cdef OBIDMS_column id1_col
cdef OBIDMS_column_p id1_col_p
cdef OBIDMS_column_p* id1_col_pp
cdef OBIDMS_column id2_col
cdef OBIDMS_column_p id2_col_p
cdef OBIDMS_column_p* id2_col_pp
cdef OBIDMS_column ocol
cdef OBIDMS_column_p ocol_p
cdef OBIDMS_column_p* ocol_pp
cdef str id1_col_name
cdef str id2_col_name
cdef str score_col_name
id1_col_name = "ID1" # TODO discuss names, aliases
id2_col_name = "ID2"
score_col_name = "score"
iview1= self
iview1_p = iview1.pointer
icol1 = iview1[bytes2str(NUC_SEQUENCE_COLUMN)]
icol1_pp = icol1.pointer
icol1_p = icol1_pp[0]
# Create the output view if needed
if output_view is None :
oview = self.dms.new_view("alignment_score_view") # TODO discuss
elif type(output_view) == str :
oview = self.dms.new_view(output_view)
else :
oview = output_view
oview.add_column(id1_col_name, type='OBI_STR', create=True)
oview.add_column(id2_col_name, type='OBI_STR', create=True)
oview.add_column(score_col_name, type='OBI_FLOAT', create=True)
oview_p = oview.pointer
ocol = oview[score_col_name]
ocol_pp = ocol.pointer
ocol_p = ocol_pp[0]
id1_col = oview[id1_col_name]
id2_col = oview[id2_col_name]
id1_col_pp = id1_col.pointer
id2_col_pp = id2_col.pointer
id1_col_p = id1_col_pp[0]
id2_col_p = id2_col_pp[0]
if obi_align_one_column(iview1_p, icol1_p, oview_p, id1_col_p, id2_col_p, ocol_p, threshold, normalize, reference, similarity_mode) < 0 :
raise Exception("Error aligning sequences")
#############################################

11
python/obitools3/obidms/_obidmscolumn_seq.pxd
View File

@ -8,17 +8,6 @@ cdef class OBIDMS_column_seq(OBIDMS_column):
cpdef object get_line(self, index_t line_nb)
cpdef set_line(self, index_t line_nb, object value)
# TO DISCUSS :
# I'am not sure that this method has to be declared here
# Alignment must be declared outside of the sequence object
cpdef align(self,
OBIView score_view,
OBIDMS_column score_column,
double threshold = *,
bint normalize = *,
int reference = *,
bint similarity_mode = *)
cdef class OBIDMS_column_multi_elts_seq(OBIDMS_column_multi_elts):
cpdef object get_item(self, index_t line_nb, str element_name)

12
python/obitools3/obidms/_obidmscolumn_seq.pyx
View File

@ -37,18 +37,6 @@ cdef class OBIDMS_column_seq(OBIDMS_column):
else :
if obi_set_seq_with_elt_idx_and_col_p_in_view(self.view.pointer, (self.pointer)[0], line_nb, 0, str2bytes(value)) < 0:
raise Exception("Problem setting a value in a column")
# TODO choose alignment type (lcs or other) with supplementary argument
cpdef align(self,
OBIView score_view,
OBIDMS_column score_column,
double threshold = 0.0,
bint normalize = True,
int reference = 0, # TODO
bint similarity_mode = True):
if (obi_align_one_column(self.view.pointer, (self.pointer)[0], score_view.pointer, (score_column.pointer)[0], threshold, normalize, reference, similarity_mode) < 0) :
raise Exception("An error occurred while aligning sequences")
cdef class OBIDMS_column_multi_elts_seq(OBIDMS_column_multi_elts):

11
python/obitools3/obidms/capi/obialign.pxd
View File

@ -6,5 +6,14 @@ from ..capi.obidmscolumn cimport OBIDMS_column_p
cdef extern from "obi_align.h" nogil:
int obi_align_one_column(Obiview_p seq_view, OBIDMS_column_p seq_column, Obiview_p score_view, OBIDMS_column_p score_column, double threshold, bint normalize, int reference, bint similarity_mode)
int obi_align_one_column(Obiview_p seq_view,
OBIDMS_column_p seq_column,
Obiview_p score_view,
OBIDMS_column_p id1_column,
OBIDMS_column_p id2_column,
OBIDMS_column_p score_column,
double threshold,
bint normalize,
int reference,
bint similarity_mode)