"""Observe extraction/filter outputs; the reference implementation is oracle.py.""" import json,sys from pathlib import Path from intronIC.extraction.annotator import AnnotationHierarchyBuilder from intronIC.extraction.intronator import IntronGenerator from intronIC.extraction.sequences import SequenceExtractor from intronIC.extraction.filters import IntronFilter from intronIC.file_io.indexed_genome import IndexedGenomeReader fa,gff=sys.argv[1:] rows=[] for mode in ['cached','indexed']: builder=AnnotationHierarchyBuilder(['exon']) genes=builder.build_from_file(gff) introns=list(IntronGenerator().generate_from_genes(genes,builder.feature_index)) if mode=='cached': extractor=SequenceExtractor(fa) else: reader=IndexedGenomeReader(fa) extractor=SequenceExtractor.from_indexed_reader(fa,reader) items=list(extractor.extract_sequences(introns,flank_size=100)) for i in IntronFilter(min_length=30,allow_noncanonical=True).filter_introns(items): rows.append({'mode':mode,'chrom':i.coordinates.chromosome,'body':i.sequences.seq,'five':i.sequences.five_seq,'three':i.sequences.three_seq,'bp':i.sequences.bp_region_seq,'omitted':i.metadata.omitted.name}) if mode=='indexed': reader.close() print(json.dumps(rows,indent=2))