def check_feature_limits(self):
if self.f.location.start > self.scaffold_size:
self.wa_errors.append(WAError(WAError.OUTSIDE_SCAFFOLD_START,
self))
if self.f.location.end > self.scaffold_size:
self.wa_errors.append(WAError(WAError.OUTSIDE_SCAFFOLD_END, self))
fmax = None
fmin = None
buggy_strand = False
for child in self.f.sub_features:
if child.type == "mRNA":
if not fmin or child.location.start < fmin:
fmin = child.location.start
if not fmax or child.location.end > fmax:
fmax = child.location.end
if not buggy_strand and child.location.strand != self.f.location.strand:
self.wa_errors.append(
WAError(WAError.WRONG_GENE_STRAND, self,
{'expected': child.location.strand}))
buggy_strand = True # Don't report for each mrna
if fmin and fmin > self.f.location.start:
self.wa_errors.append(
WAError(WAError.WRONG_GENE_START, self, {'expected': fmin}))
if fmax and fmax < self.f.location.end:
self.wa_errors.append(
WAError(WAError.WRONG_GENE_END, self, {'expected': fmax}))
def check_sub_features_mrna(self):
for child in self.f.sub_features:
if child.type != "mRNA":
self.wa_errors.append(
WAError(
WAError.UNEXPECTED_SUB_FEATURE, self, {
'child_id': child.qualifiers['ID'][0],
'child_type': child.type
}))
def check_sub_features(self):
if len(self.f.sub_features) > 1:
foundOverlap = False
for sub1 in self.f.sub_features:
firstChild = sub1
start1 = firstChild.location.start
end1 = firstChild.location.end
for sub2 in sub1.sub_features:
if (sub1.qualifiers['ID'][0] != sub2.qualifiers['ID'][0]):
start2 = sub2.location.start
end2 = sub2.location.end
if ((start2 >= start1 and start2 <= end1)
or (end2 >= start1 and end2 <= end1)
or (start1 >= start2 and start1 <= end2)
or (end1 >= start2 and end1 <= end2)):
foundOverlap = True
if not foundOverlap:
self.wa_errors.append(
WAError(WAError.MULTIPLE_SUB_FEATURE, self,
{'num_children': len(self.f.sub_features)}))
else:
self.check_multiple_mrnas()
def validate_genes(self):
in_handle = open(self.in_file)
for rec in GFF.parse(in_handle):
for f in rec.features:
if (f.type == "gene") and (
'status' not in f.qualifiers
or not f.qualifiers['status']
or f.qualifiers['status'][0].lower() != "deleted"):
gene = Gene(f, rec.id, self.scaf_lengths[rec.id],
self.allowed_groups, self.group_tags,
self.no_group, self.split_users)
self.all_genes[gene.wa_id] = gene
# Count number of genes with goid
if gene.has_goid:
self.genes_with_goid += 1
# Collect stats on groups
for g in gene.groups:
if g not in self.groups_stats:
self.groups_stats[g] = 0
self.groups_stats[g] += 1
new_part = gene.part
new_allele = gene.allele
# Collect wa_errors
self.wa_errors.extend(gene.wa_errors)
if not new_part and not new_allele:
self.genes_seen_once += 1
# keep track of splitted genes
if new_part:
part_gene_key = gene.display_id
if new_allele:
part_gene_key = gene.display_id + ", allele " + new_allele
if part_gene_key not in self.splitted_genes:
self.splitted_genes[part_gene_key] = {}
if new_part in self.splitted_genes[part_gene_key]:
identical = self.splitted_genes[part_gene_key][
new_part]
gene.errors.append(
GeneError(
GeneError.PART_SAME, gene, {
'other_name': identical.display_id,
'other_scaff': identical.scaffold,
'other_start':
identical.f.location.start,
'other_end': identical.f.location.end
}))
self.splitted_genes[part_gene_key][new_part] = gene
# keep track of duplicated genes
if new_allele:
allele_gene_key = gene.display_id
if allele_gene_key not in self.duplicated_genes:
self.duplicated_genes[allele_gene_key] = {}
if new_allele in self.duplicated_genes[
allele_gene_key]:
identical = self.duplicated_genes[allele_gene_key][
new_allele]
if identical.part == new_part:
gene.errors.append(
GeneError(
GeneError.ALLELE_SAME, gene, {
'other_name': identical.display_id,
'other_scaff': identical.scaffold,
'other_start':
identical.f.location.start,
'other_end':
identical.f.location.end
}))
self.duplicated_genes[allele_gene_key][
new_allele] = gene
elif 'status' in f.qualifiers and f.qualifiers[
'status'] and f.qualifiers['status'][0].lower(
) == "deleted":
gene = Gene(f, rec.id, self.scaf_lengths[rec.id],
self.allowed_groups, self.group_tags,
self.no_group, self.split_users)
self.all_genes[gene.wa_id] = gene
else:
fake_gene = Gene(f, rec.id, self.scaf_lengths[rec.id],
self.allowed_groups, self.group_tags)
self.wa_errors.append(
WAError(WAError.UNEXPECTED_FEATURE, fake_gene))
in_handle.close()