def execute_mat2mat(self):
matrix_list = self.create_matrix_list()
sv = None
if self.sort_vector:
sv = self.sort_vector.vector.path
mm = MatrixByMatrix(
matrix_list=matrix_list,
window_start=self.bin_start,
bin_number=self.bin_number,
bin_size=self.bin_size,
sort_vector=sv,
)
fn = get_random_filename(os.path.join(settings.MEDIA_ROOT, self.UPLOAD_TO))
mm.writeJson(fn)
return os.path.join(self.UPLOAD_TO, os.path.basename(fn))
def execute(cls, analysis, dataset):
# returns a new or existing FeatureListCountMatrix that matches the
# specified criteria
# Find existing instance
existing = cls.objects.filter(
feature_list=analysis.feature_list,
dataset=dataset,
anchor=analysis.anchor,
bin_start=analysis.bin_start,
bin_number=analysis.bin_number,
bin_size=analysis.bin_size,
).first()
if existing:
return existing
# existing not found; create instead
fn = get_random_filename(os.path.join(settings.MEDIA_ROOT, cls.UPLOAD_TO))
bigwigs = dataset.get_bigwig_paths()
BedMatrix(
bigwigs=bigwigs,
feature_bed=analysis.feature_list.dataset.path,
output_matrix=fn,
anchor=analysis.get_anchor_display(),
bin_start=analysis.bin_start,
bin_number=analysis.bin_number,
bin_size=analysis.bin_size,
opposite_strand_fn=None,
stranded_bigwigs=dataset.is_stranded,
stranded_bed=analysis.feature_list.stranded
)
return cls.objects.create(
feature_list=analysis.feature_list,
dataset=dataset,
anchor=analysis.anchor,
bin_start=analysis.bin_start,
bin_number=analysis.bin_number,
bin_size=analysis.bin_size,
matrix=os.path.join(cls.UPLOAD_TO, os.path.basename(fn))
)
