6a6d7a78f1de5456925fc8f4f675795cbb61ee35 max Tue Nov 29 03:16:02 2022 -0800 adding tests for bigGuessDb, refs #30316 diff --git src/utils/bigGuessDb/bigGuessDb src/utils/bigGuessDb/bigGuessDb index 419b62a..6e9241a 100755 --- src/utils/bigGuessDb/bigGuessDb +++ src/utils/bigGuessDb/bigGuessDb @@ -6,82 +6,88 @@ import logging, optparse, sys from collections import defaultdict from os.path import join, isfile, expanduser import os, gzip, subprocess # ==== functions ===== def parseArgs(): " setup logging, parse command line arguments and options. -h shows auto-generated help page " parser = optparse.OptionParser("usage: %prog [options] filename - given a bigBed or "\ "bigWig file, " \ "guess the assembly based on the chrom sizes") parser.add_option("-d", "--debug", dest="debug", action="store_true", \ help="show debug messages") parser.add_option("", "--index", dest="index", action="store_true", \ - help="go /hive/data/genomes and build a file ~/.guessAssembly.txt with all chromSizes") + help="go /hive/data/genomes and build an index file with all chromSizes") parser.add_option("-b", "--best", dest="best", action="store_true", \ - help="only print a single string, the best matching assembly") + help="only print a single string, the best matching assembly, or 'emptyFile' or 'notFound'. ") + parser.add_option("", "--indexFile", dest="indexFname", action="store", \ + help="Use specified index file, default is %default", default="~/.guessAssembly.txt.gz") (options, args) = parser.parse_args() if args==[] and not options.index: parser.print_help() exit(1) if options.debug: logging.basicConfig(level=logging.DEBUG) logging.getLogger().setLevel(logging.DEBUG) else: logging.basicConfig(level=logging.INFO) logging.getLogger().setLevel(logging.INFO) return args, options def parseSizes(inFname, doSubset): " given a chrom.sizes file, return the 10 longest and 10 shortest chrom names " logging.info("Reading %s",inFname) sizes = list() for line in open(inFname): if line.startswith("#"): continue chrom, size = line.rstrip("\n").split("\t")[:2] sizes.append( (int(size), chrom) ) + + if len(sizes)==0: + logging.error("bigBed/bigWig file is empty. Cannot guess assembly.") + return None + sizes.sort() if not doSubset: return sizes - someSizes = sizes[:10] - someSizes.extend(sizes[-10:]) + someSizes = sizes[-20:] # small chroms carry less information and have fewer features return someSizes def writeSizes(allSizes, outFname): " write all sizes to the index file " ofh = gzip.open(outFname, "wt") # "write" "text" for db, dbSizes in allSizes.items(): sizeParts = ["%s=%d" % (chrom, size) for size,chrom in dbSizes] sizeStr = ",".join(sizeParts) ofh.write("%s\t%s\n" % (db, sizeStr)) ofh.close() logging.info("Wrote %s", outFname) def buildIndex(inDir, outFname): """ go over all direct subdirectories of inDir and find a chrom.sizes file, compact it to format db -> list of (chrom,size) and write to outFname """ allSizes = dict() for db in os.listdir(inDir): - if "Patch" in db or db == "sonMus0" or db.startswith("braNey"): + if "Patch" in db or db == "sonMus0" or db.startswith("braNey") or db.endswith(".lowec"): continue subDir = join(inDir, db) chromFname = join(subDir, "chrom.sizes") if not isfile(chromFname): chromFname = join(subDir, db+".sizes") if not isfile(chromFname): print("not found "+chromFname) continue doSubset = True if db.startswith("hg") or db.startswith("mm"): doSubset = False @@ -155,30 +161,39 @@ dbMatches = list(dbChromMatch.items()) # dbMatches is now a list of db -> list of chromosomes dbMatches.sort(key=sortBySecondLen) return dbMatches def printAllMatches(dbChromMatch): " print all matching dbs as a tsv " print("#db\tmatchCount\tmatchList") for db, chromList in dbChromMatch: print("\t".join([db, str(len(chromList)), ",".join(chromList)])) # ----------- main -------------- def main(): " entry point to script " args, options = parseArgs() - indexFname = expanduser("~/.guessAssembly.txt.gz") + indexFname = expanduser(options.indexFname) if options.index: buildIndex("/hive/data/genomes", indexFname) else: inFname = args[0] fileSizes = bigSizes(inFname) + if len(fileSizes) == 0: + logging.debug("%s is empty. Cannot determine assembly." % inFname) + hits = ( ("emptyFile", 0), ) + else: sizeIndex = readSizeIndex(indexFname) hits = findBestDb(sizeIndex, fileSizes) + if options.best: + if len(hits)==0: + print("notFound") + else: + #if (len(hits[0][1]) >= 2): # need more than a single match, as chrM often matches print(hits[0][0]) else: printAllMatches(hits) main()