def startwf1(indir,resultdir,checkpoint=False,concat=False,mashmxdist=0.5,cpu=1,
skip="",refdb="",hmmdb="",rnadb="",maxmlst=100,model="GTR",bs=0,
kf=False,maxorg=50,filtMLST=True,fast=False,minmlst=10):
"""WORKFLOW 1: Build phylogeny from scratch"""
if not checkpoint:
checkpoint = "w1-0"
queryseqs = os.path.join(resultdir,"queryseqs")
#Parse all inputs
if checkpoint == "w1-0":
log.info("JOB_STATUS::Parsing all genomes...")
if parsegenomes.parseall(indir,queryseqs):
checkpoint = "w1-1"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::5/100")
else:
log.error("Problem parsing input genomes")
return False
#Run MASH distances
mashresult = False
if checkpoint == "w1-1":
log.info("JOB_STATUS::Running MASH ANI estimation against reference sequences...")
mashresult = mash.getdistances(queryseqs,resultdir,cpu=cpu,maxdist=mashmxdist)
if mashresult:
checkpoint = "w1-2"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::10/100")
else:
log.error("MASH distance failed")
return False
#Get set of organisms to build seq DB
selorgs = False
if checkpoint == "w1-2":
log.info("JOB_STATUS::Loading mash results...")
if mashresult:
pass
elif os.path.exists(os.path.join(resultdir,"reflist.json")):
with open(os.path.join(resultdir,"reflist.json"),"r") as fil:
mashresult = json.load(fil)
log.info("Loading mash results...")
else:
log.error("No Mash results to process")
return False
selorgs = getorgs(resultdir,mashresult,skip=skip,IGlimit=maxorg,minorgs=25)
if selorgs:
checkpoint = "w1-3"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::15/100")
else:
checkpoint = "w1-STEP2"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
if checkpoint == "w1-STEP2":
log.info("JOB_STATUS:: Waiting for selected organisms")
return "waiting"
#Copy reference sequence database and add query organisms
orgdb = os.path.join(resultdir,"refquery.db")
# selorgs = {}
if not selorgs and os.path.exists(os.path.join(resultdir,"userlist.json")):
with open(os.path.join(resultdir,"userlist.json"),"r") as fil:
selorgs = json.load(fil)
elif not selorgs and os.path.exists(os.path.join(resultdir,"autoOrglist.json")):
with open(os.path.join(resultdir,"autoOrglist.json"),"r") as fil:
selorgs = json.load(fil)
if checkpoint == "w1-3":
log.info("JOB_STATUS:: Collating selected genomes...")
#Clear old db if exists
if os.path.exists(orgdb):
os.remove(orgdb)
flist = list(selorgs["selspecies"])
flist.extend(selorgs["seloutgroups"])
flist = [x for x in flist if "query" not in x.lower()]
log.info("refdb: %s"%refdb)
if copyseqsql.copydb(flist,refdb,orgdb):
#get file list of query orgs and add to db
seqlist = glob.glob(os.path.join(queryseqs,"*.fna"))
allorgs = makeseqsql.runlist(seqlist,orgdb,True,"",True)
if allorgs:
checkpoint = "w1-4"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
else:
log.error("Failed at copydb")
#Run HMM searches on query organism
if checkpoint == "w1-4":
if not orgdb:
log.error("No querydb rerun at checkpoint w1-3")
return False
#Write newly added seqences to file
naseqs = os.path.join(resultdir,"addedseqs.fna")
aaseqs = os.path.join(resultdir,"addedseqs.faa")
seqsql2fa.writefasta(orgdb,aaseqs,False,"",True)
seqsql2fa.writefasta(orgdb,naseqs,True,"",True)
#Run HMM searches
log.info("JOB_STATUS:: Searching for MLST genes in query sequences...")
log.info("JOB_PROGRESS::25/100")
if not hmmdb:
hmmdb = os.path.join(os.path.dirname(os.path.realpath(__file__)),"reducedcore.hmm")
if not rnadb:
rnadb = os.path.join(os.path.dirname(os.path.realpath(__file__)),"barnap_bact_rRna.hmm")
hmmsearch(aaseqs+".domhr",hmmdb,aaseqs,mcpu=cpu)
hmmsearch(naseqs+".domhr",rnadb,naseqs,mcpu=cpu)
#Add HMMresults to refdb and cleanup unused seqs
log.info("Adding query HMM results to database")
status = makehmmsql.run(aaseqs+".domhr",orgdb)
if not status:
#If no genes were found report error
log.error("No MLST genes could be found. Stop processing")
log.info("Adding query RNA results to database")
status = makehmmsql.run(naseqs+".domhr",orgdb,rna=True)
os.remove(aaseqs)
os.remove(naseqs)
checkpoint = "w1-5"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
#Calculate mlst core genes and export
mlstdir = os.path.join(resultdir,"mlstgenes")
mlstpriority = os.path.join(resultdir,"mlstpriority.json")
genematjson = os.path.join(resultdir,"mlstmatrix.json")
dndsfile = os.path.join(os.path.dirname(os.path.realpath(__file__)),"dnds.json")
mlstselection = []
delorgs = []
if checkpoint == "w1-5":
allquery = [os.path.splitext(os.path.split(x)[-1])[0].replace(" ","_") for x in glob.glob(os.path.join(resultdir,"queryseqs","*.fna"))]
ignoreorgs = list(selorgs.get("seloutgroups",[]))
ignoreorgs.extend(allquery)
#getmlstgenes.findsingles(orgdb,maxgenes=500,outdir=mlstdir)
if os.path.exists(mlstpriority) and os.path.exists(genematjson):
with open(genematjson,"r") as mfil, open(mlstpriority,"r") as pfil:
mlstpriority = json.load(pfil)
temp = json.load(mfil)
genemat = temp["counts"]
orgs = temp["orgs"]
del temp
else:
genemat,orgs,mlstpriority = getgenematrix.getmat(orgdb,pct=0.5,pct2=1.0,bh=True,rna=True,savefil=genematjson,prifile=mlstpriority,dndsfile=dndsfile,ignoreorgs=allquery)
mlstselection, delorgs, concat = getmlstselection(resultdir,mlstpriority,maxmlst,ignoreorgs=ignoreorgs,concat=concat,minmlst=minmlst)
if "skip3" in skip.lower() or os.path.exists(os.path.join(resultdir,"usergenes.json")):
#Export selected genes to mlst folder
log.info("JOB_STATUS:: Writing MLST genes...")
getgenes.writeallgenes(orgdb,mlstselection,delorgs,outdir=mlstdir,outgroups=selorgs.get("seloutgroups",None),pct=0.5,rename=True)
checkpoint = "w1-6"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
else:
checkpoint = "w1-STEP3"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
if checkpoint == "w1-STEP3":
log.info("JOB_STATUS:: Waiting for selected MLST genes")
return "waiting"
## Align and trim all MLST genes
aligndir = os.path.join(resultdir,"mlst_aligned")
trimdir = os.path.join(resultdir,"mlst_trimmed")
if checkpoint == "w1-6":
log.info("JOB_STATUS:: Aligning MLST genes")
log.info("JOB_PROGRESS::30/100")
#align all
processmlst(mlstdir,aligndir,cpu=cpu,fast=fast)
checkpoint = "w1-6b"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
if checkpoint == "w1-6b":
if filtMLST and filtMLST != "False":
#Extra screen of MLST genes to remove outliers based on starting tree distance
log.info("JOB_STATUS:: Screening for inconsistent MLST genes")
log.info("JOB_PROGRESS::55/100")
excludemlst = screenmlst(mlstdir,aligndir,cpu=cpu)
log.info("JOB_STATUS:: Excluded MLST genes: %s"%excludemlst)
#trim all
log.info("JOB_STATUS:: Trimming alignments")
log.info("JOB_PROGRESS::65/100")
processmlst(aligndir,trimdir,cpu=cpu,trim=True)
checkpoint = "w1-7"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
treedir = os.path.join(resultdir,"trees")
finishedtree = ""
#Build trees
if checkpoint == "w1-7":
log.info("JOB_PROGRESS::75/100")
if concat and concat != "False":
log.info("JOB_STATUS:: Running concatenated supermatrix phylogeny")
concatfasta = os.path.join(resultdir,"concatMLST.fasta")
partfile = os.path.join(resultdir,"nucpartition.txt")
concatphylogeny(resultdir, concatfasta, partfile,cpu=cpu,model=model,bs=bs)
checkpoint = "w1-F"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
else:
log.info("JOB_STATUS:: Running coalescent tree phylogeny")
colphylogeny(resultdir,trimdir,cpu=cpu,model=model,bs=bs)
checkpoint = "w1-F"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
if checkpoint == "w1-F":
finaltree = os.path.join(resultdir,"final.tree")
if os.path.exists(os.path.join(treedir,"concatTree.tree.treefile")) and concat:
finishedtree = os.path.join(treedir,"concatTree.tree.treefile")
elif os.path.exists(os.path.join(treedir,"summaryTree.tree")):
finishedtree = os.path.join(treedir,"summaryTree.tree")
#Copy final tree to root dir
if finishedtree:
fmat = 2 if bs or not concat else 5
log.debug("Saving final tree... %s"%fmat)
if not ete3helper.rerootTree(finishedtree,finaltree,fmat=fmat):
shutil.copy(finishedtree,finaltree)
else:
log.error("Could not find final tree. Tree building failed")
#Job finished? do some cleanup
if not kf:
if os.path.exists(orgdb):
os.remove(orgdb)
shutil.rmtree(os.path.join(resultdir,"queryseqs"))
shutil.rmtree(os.path.join(resultdir,"mlst_trimmed"))
shutil.rmtree(os.path.join(resultdir,"mlstgenes"))
for oldfil in glob.glob(os.path.join(treedir,"*.model")):
os.remove(oldfil)
def startwf2(indir,resultdir,refdir="",checkpoint=False,reference="",model="GTR",bs=0,kf=False,maxmlst=100,
cpu=1,mashmxdist=0.5,rnadb="",hmmdb="",fast=False):
"""WORKFLOW 2: Get all query genomes and identify reference tree to add sequences to"""
if not checkpoint:
checkpoint = "w2-0"
if reference:
checkpoint = "w2-3"
queryseqs = os.path.join(resultdir,"queryseqs")
#Parse all inputs
if checkpoint == "w2-0":
log.info("JOB_STATUS::Parsing all genomes...")
if parsegenomes.parseall(indir,queryseqs):
checkpoint = "w2-1"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::5/100")
else:
log.error("Problem parsing input genomes")
return False
#Run MASH distances
mashresult = False
if checkpoint == "w2-1":
log.info("JOB_STATUS::Running MASH ANI estimation against reference sequences...")
mashresult = mash.getdistances(queryseqs,resultdir,cpu=cpu,maxdist=mashmxdist)
if mashresult:
checkpoint = "w2-2"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::10/100")
else:
log.error("MASH distance failed")
return False
#Detect reference group
if checkpoint == "w2-2":
log.info("JOB_STATUS::Loading mash results...")
if mashresult:
pass
elif os.path.exists(os.path.join(resultdir,"reflist.json")):
with open(os.path.join(resultdir,"reflist.json"),"r") as fil:
mashresult = json.load(fil)
log.info("Loading mash results...")
else:
log.error("No Mash results to process")
return False
reference = getreference(mashresult,refdir)
if reference:
checkpoint = "w2-3"
log.info("JOB_STATUS::Detected reference = %s"%reference)
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.info("JOB_PROGRESS::15/100")
else:
checkpoint = "w2-F"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
log.error("No matching reference found for all query organisms")
orgdb = os.path.join(resultdir,"allseqs.db")
seqlist = glob.glob(os.path.join(queryseqs,"*.fna"))
#Add to db and run HMM searches. Use model in reference folder, or use global models
if checkpoint == "w2-3":
#get file list of query orgs and add to db
allorgs = makeseqsql.runlist(seqlist,orgdb,True,"",False)
#Write newly added seqences to file
naseqs = os.path.join(resultdir,"addedseqs.fna")
aaseqs = os.path.join(resultdir,"addedseqs.faa")
seqsql2fa.writefasta(orgdb,aaseqs,False,"",True)
seqsql2fa.writefasta(orgdb,naseqs,True,"",True)
#Run HMM searches
log.info("JOB_STATUS:: Searching for MLST genes in query sequences...")
log.info("JOB_PROGRESS::25/100")
if not hmmdb:
if os.path.exists(os.path.join(refdir,reference,"core.hmm")):
hmmdb = os.path.join(refdir,reference,"core.hmm")
else:
hmmdb = os.path.join(os.path.dirname(os.path.realpath(__file__)),"reducedcore.hmm")
if not rnadb:
rnadb = os.path.join(os.path.dirname(os.path.realpath(__file__)),"barnap_bact_rRna.hmm")
hmmsearch(aaseqs+".domhr",hmmdb,aaseqs,mcpu=cpu)
hmmsearch(naseqs+".domhr",rnadb,naseqs,mcpu=cpu)
#Add HMMresults
log.info("Adding query HMM results to database")
status = makehmmsql.run(aaseqs+".domhr",orgdb)
if not status:
#If no genes were found report error
log.error("No MLST genes could be found. Stop processing")
log.info("Adding query RNA results to database")
status = makehmmsql.run(naseqs+".domhr",orgdb,rna=True)
os.remove(aaseqs)
os.remove(naseqs)
checkpoint = "w2-4"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
mlstdir = os.path.realpath(os.path.join(resultdir,"mlstgenes"))
if not os.path.exists(mlstdir):
os.makedirs(mlstdir)
aligndir = os.path.realpath(os.path.join(resultdir,"mlst_aligned"))
if not os.path.exists(aligndir):
os.makedirs(aligndir)
trimdir = os.path.realpath(os.path.join(resultdir,"mlst_trimmed"))
if not os.path.exists(trimdir):
os.makedirs(trimdir)
# extract single copy results from genelist
if checkpoint == "w2-4":
genelist = [os.path.splitext(os.path.split(x)[1])[0] for x in glob.glob(os.path.join(refdir,reference,"*.fna"))]
log.info("JOB_STATUS:: Writing MLST genes...")
singles = [str(x) for x in getgenematrix.getmat(orgdb,rna=True,bh=True)[2][0] if x in genelist]
getgenes.writeallgenes(orgdb,singles,[],outdir=mlstdir,outgroups=None,pct=0.5,writeaa=False,rename=True)
log.info("JOB_STATUS:: Adding MLST genes to alignments...")
addallalign([os.path.join(mlstdir,x+".fna") for x in singles], os.path.join(refdir,reference), aligndir, cpu=cpu,fast=fast)
log.info("JOB_STATUS:: Trimming all alignments...")
processmlst(aligndir,trimdir,cpu=cpu,trim=True)
checkpoint = "w2-5"
log.info("JOB_CHECKPOINT::%s"%checkpoint)
treedir = os.path.realpath(os.path.join(resultdir,"trees"))
if not os.path.exists(treedir):
os.makedirs(treedir)
#Add to prebuilt trees
if checkpoint == "w2-5":
log.info("JOB_PROGRESS::55/100")
log.info("JOB_STATUS:: Adding to reference gene trees...")
inlist = [os.path.join(trimdir,x) for x in os.listdir(trimdir)]
addalltrees(inlist,os.path.join(refdir,reference),treedir,cpu=cpu)
checkpoint = "w2-6"
log.info("JOB_CHECKPOINT::%s" % checkpoint)
if checkpoint == "w2-6":
log.info("JOB_PROGRESS::85/100")
log.info("JOB_STATUS:: Running coalescent tree phylogeny")
#Combine all trees using ASTRAL
flist = glob.glob(os.path.join(treedir,"RAxML_labelledTree.*.tree"))
#reformat newick trees for astral
for pbt in flist:
with open(pbt,"r") as ifil, open(pbt+".newick","w") as ofil:
x = ifil.next()
ofil.write(re.sub("\[I\d+?\]|\"|'|QUERY___","",x))
flist = glob.glob(os.path.join(treedir,"*.newick"))
alltrees = catTrees(flist,os.path.join(treedir,"alltrees.tree"))
coltree = os.path.join(treedir,"summaryTree.tree")
runAstral(resultdir,alltrees,coltree)
checkpoint = "w2-F"
log.info("JOB_CHECKPOINT::%s" % checkpoint)
finishedtree=""
if checkpoint == "w2-F":
finaltree = os.path.join(resultdir,"final.tree")
if os.path.exists(os.path.join(treedir,"summaryTree.tree")):
finishedtree = os.path.join(treedir,"summaryTree.tree")
#Copy final tree to root dir
if finishedtree:
fmat = 2
log.debug("Saving final tree... %s"%fmat)
if not ete3helper.rerootTree(finishedtree,finaltree,fmat=fmat):
shutil.copy(finishedtree,finaltree)
else:
log.error("Could not find final tree. Tree building failed")
def startUPLBwf(indir,
resultdir,
checkpoint=False,
concat=False,
mashmxdist=0.5,
cpu=1,
skip="",
refdb="",
hmmdb="",
rnadb="",
maxmlst=100,
model="GTR",
bs=0,
kf=False,
maxorg=50,
filtMLST=True,
fast=False,
minmlst=10):
"""WORKFLOW 1: Build phylogeny from scratch"""
if not checkpoint:
checkpoint = "w1-0"
queryseqs = os.path.join(resultdir, "queryseqs")
#Parse all inputs
if checkpoint == "w1-0":
log.info("JOB_STATUS::Parsing all genomes...")
if parsegenomes.parseall(indir, queryseqs):
checkpoint = "w1-1"
log.info("JOB_CHECKPOINT::%s" % checkpoint)
log.info("JOB_PROGRESS::5/100")
else:
log.error("Problem parsing input genomes")
return False
#Run MASH distances
mashresults = False
if checkpoint == "w1-1":
for idx, queryseq in enumerate(queryseqs):
log.info(
"JOB_STATUS::Running MASH ANI estimation against reference sequences...{}"
.format(len(queryseqs)))
mashresult = mash.getdistances([queryseq],
resultdir,
cpu=cpu,
maxdist=mashmxdist)
if mashresult:
# checkpoint = "w1-2"
log.info("JOB_CHECKPOINT::%s" % checkpoint)
log.info("JOB_PROGRESS::Seq {} of {}".format(
idx + 1, len(queryseqs)))
mashresults.append(mashresult)
else:
log.error("MASH distance failed")
return False
#Get set of organisms to build seq DB
selorgDict = False
if checkpoint == "w1-2":
log.info("JOB_STATUS::Loading mash results...")
for mashresult in mashresults:
selorgs = getorgs(resultdir,
mashresult,
skip=skip,
IGlimit=maxorg,
minorgs=25)
selorgDict = zip(queryseqs, )
with open(autosel, "w") as fil:
json.dump(selection, fil)
print(selorgs)