def testGetNumOfBins(self): self.assertEqual(0, CompBinManager.getNumOfBins(GenomeRegion('hg18', 'chr1', 0, 0))) self.assertEqual(1, CompBinManager.getNumOfBins(GenomeRegion('hg18', 'chr1', 0, 100))) self.assertEqual(2, CompBinManager.getNumOfBins(GenomeRegion('hg18', 'chr1', 200, 400))) self.assertEqual(4, CompBinManager.getNumOfBins(GenomeRegion('hg18', 'chr1', 67, 314)))
def storeBoundingRegions(self, boundingRegionTuples, genomeElementChrList, sparse): assert sparse in [False, True] tempContents = OrderedDict() genomeElementChrs = set(genomeElementChrList) lastRegion = None chrStartIdxs = OrderedDict() chrEndIdxs = OrderedDict() totElCount = 0 totBinCount = 0 for br in boundingRegionTuples: if lastRegion is None or br.region.chr != lastRegion.chr: if br.region.chr in tempContents: raise InvalidFormatError("Error: bounding region (%s) is not grouped with previous bounding regions of the same chromosome (sequence)." % br.region) lastRegion = None tempContents[br.region.chr] = OrderedDict() if sparse: chrStartIdxs[br.region.chr] = totElCount else: if br.region < lastRegion: raise InvalidFormatError("Error: bounding regions in the same chromosome (sequence) are unsorted: %s > %s." % (lastRegion, br.region)) if lastRegion.overlaps(br.region): raise InvalidFormatError("Error: bounding regions '%s' and '%s' overlap." % (lastRegion, br.region)) if lastRegion.end == br.region.start: raise InvalidFormatError("Error: bounding regions '%s' and '%s' are adjoining (there is no gap between them)." % (lastRegion, br.region)) if len(br.region) < 1: raise InvalidFormatError("Error: bounding region '%s' does not have positive length." % br.region) if not sparse and len(br.region) != br.elCount: raise InvalidFormatError("Error: track type representation is dense, but the length of bounding region '%s' is not equal to the element count: %s != %s" % (br.region, len(br.region), br.elCount)) startIdx, endIdx = (totElCount, totElCount + br.elCount) if not sparse else (None, None) totElCount += br.elCount if sparse: chrEndIdxs[br.region.chr] = totElCount tempContents[br.region.chr][br.region.start] = BoundingRegionInfo(br.region.start, br.region.end, startIdx, endIdx, 0, 0) lastRegion = br.region if sparse: totBinCount = 0 for chr in tempContents: chrLen = GenomeInfo.getChrLen(self._genome, chr) numBinsInChr = CompBinManager.getNumOfBins(GenomeRegion(start=0, end=chrLen)) for key in tempContents[chr].keys(): startBinIdx = totBinCount endBinIdx = totBinCount + numBinsInChr brInfo = tempContents[chr][key] if chr in genomeElementChrs: tempContents[chr][key] = BoundingRegionInfo(brInfo.start, brInfo.end, \ chrStartIdxs[chr], chrEndIdxs[chr], \ startBinIdx, endBinIdx) else: if chrEndIdxs[chr] - chrStartIdxs[chr] > 0: raise InvalidFormatError("Error: bounding region '%s' has incorrect element count: %s > 0" % (GenomeRegion(chr=chr, start=brInfo.start, end=brInfo.end), chrEndIdxs[chr] - chrStartIdxs[chr])) tempContents[chr][key] = BoundingRegionInfo(brInfo.start, brInfo.end, 0, 0, 0, 0) if chr in genomeElementChrs: totBinCount += numBinsInChr if len(genomeElementChrs - set(tempContents.keys())) > 0: raise InvalidFormatError('Error: some chromosomes (sequences) contains data, but has no bounding regions: %s' % ', '.join(genomeElementChrs - set(tempContents.keys()))) ensurePathExists(self._fn) for chr in tempContents: brInfoDict = tempContents[chr] tempContents[chr] = BrInfoHolder(tuple(brInfoDict.keys()), tuple(brInfoDict.values())) brShelve = safeshelve.open(self._fn) brShelve.update(tempContents) brShelve.close() while not self.fileExists(): from gtrackcore_memmap.application.LogSetup import logMessage logMessage("Bounding region shelve file '%s' has yet to be created" % self._fn) import time time.sleep(0.2)