def step4_SCpeakbias(conf_dict,logfile):
wlog('readin sequence from 2bit',logfile)
seq_dict = {}
inf = open(conf_dict['options']['csize'])
for line in inf:
chrm = line.split()[0]
seq_dict[chrm] = fetchseq_2bit_chrom(conf_dict['General']['twoBitToFa'],conf_dict['General']['sequence'],chrm)
inf.close()
conf_dict['results']['seqdict'] = seq_dict
wlog('scan peak level bias',logfile)
tmplog = bias_peakXcell_mat(conf_dict['General']['outname'],
conf_dict['General']['bedtools'],
conf_dict['options']['chromosome'],
conf_dict['options']['kmer'],
conf_dict['results']['biasMat'],
conf_dict['results']['seqdict'],
conf_dict['results']['finalcells'],
conf_dict['General']['datatype'],
conf_dict['options']['peakminreads'],
conf_dict['options']['peakmaxreads']
)
return conf_dict
def step1_QC_format(conf_dict, logfile):
### preparing mapping state dict
wlog('summarize reads count distribution', logfile)
chrom_reads = split_chromosome_reads(conf_dict['General']['fragments'],
conf_dict['General']['outname'],
conf_dict['options']['scATAC10x'],
conf_dict['options']['chromosome'])
if "chrM" in chrom_reads:
conf_dict['QC']["chrM_reads"] = chrom_reads["chrM"]
else:
conf_dict['QC']["chrM_reads"] = 0
chromatin_reads = 0
for chrom in chrom_reads.keys():
if chrom != "chrM":
chromatin_reads += chrom_reads[chrom]
conf_dict['QC']["chromatin_reads"] = chromatin_reads
if conf_dict['General']['mode'] == "sc":
wlog('filter high quality single cells', logfile)
filter_highQcell_results = filter_highQcell_reads(
conf_dict['General']['outname'],
int(conf_dict['options']['readcutoff']),
conf_dict['options']["usecells"])
if filter_highQcell_results == "fail":
ewlog(
'obtain < 100 high quality cell with reads >= %s.' %
(conf_dict['options']['readcutoff']), logfile)
if len(conf_dict['options']["usecells"]) == 0:
wlog('no specified cellname list inputed', logfile)
elif filter_highQcell_results[4] == "highQ":
wlog(
'obtain < 100 cell left after highQ + cellname filtering, use highQ cell only',
logfile)
wlog(
'obtain %s cells from filtering, containing %s reads' %
(filter_highQcell_results[1], filter_highQcell_results[2]),
logfile)
conf_dict['results']['finalcells'] = filter_highQcell_results[0]
conf_dict['QC']['totalcellnum'] = filter_highQcell_results[3]
conf_dict['QC']['highQcellnum'] = filter_highQcell_results[1]
conf_dict['QC']['finalusecellnum'] = len(
conf_dict['results']['finalcells'])
conf_dict['QC']['finalreadnum'] = filter_highQcell_results[2]
return conf_dict
def step2_biasMat(conf_dict, logfile):
## readin 2bit seq
### obtain bias mat
if conf_dict['options']['bias'] == "naked":
wlog('obtain pre-processed bias matrix from naked DNA data', logfile)
conf_dict['results']['biasMat'] = readBias(
conf_dict['options']['biasfile'])
elif conf_dict['QC']['chrM_reads'] < 500000:
wlog(
'chrM reads number < 500k, obtain pre-processed bias matrix from naked DNA data',
logfile)
if not os.path.isfile(conf_dict['options']['biasfile']):
ewlog("no naked DNA bias matrix, cannot estimate bias", logfile)
else:
conf_dict['results']['biasMat'] = readBias(
conf_dict['options']['biasfile'])
else:
wlog('estimate bias matrix from mtDNA(chrM) data', logfile)
conf_dict['results']['biasMatNaive'] = naive_kmerBias_chrM(
conf_dict['General']['outname'], conf_dict['General']['sequence'],
conf_dict['options']['kmer'], conf_dict['General']['twoBitToFa'],
conf_dict['General']['format'])
conf_dict['results']['biasfile'] = "%s_bias.txt" % (
conf_dict['General']['outname'])
conf_dict['results']['biasMat'] = simplex_encoding(
conf_dict['results']['biasMatNaive'],
conf_dict['results']['biasfile'])
return conf_dict
def step3_callpeak(conf_dict, logfile):
conf_dict['results'][
'peakfile'] = conf_dict['General']['outname'] + "_summitEXT.bed"
macs3callpeak = 1
if conf_dict['options']['peak'] != "NA":
conf_dict['QC']['peaknumTotal'] = extExternal(
conf_dict['options']['peak'], conf_dict['results']['peakfile'],
int(conf_dict['options']['extend']))
if conf_dict['QC']['peaknumTotal'] < 1000:
wlog(
"obtain < 1000 (%s) external inputted peaks, use macs3 to detect peaks"
% conf_dict['QC']['peaknumTotal'], logfile)
macs3callpeak = 1
else:
wlog(
"obtain %s peaks from (-p) inputted" %
conf_dict['QC']['peaknumTotal'], logfile)
macs3callpeak = 0
if macs3callpeak == 1:
if conf_dict['General']['macs3'] == "NA":
ewlog(
"macs3 was not installed. SELMA requires macs3 installed in the default path ($PATH) for peak calling",
logfile)
### callpeak
if conf_dict['General']['genome'] == "hg38":
gtag = "hs"
else:
gtag = "mm"
if conf_dict['General']['format'] == "PE":
macs3cmd = "macs3 callpeak -t %s -n %s -f BEDPE -g %s -q %s --keep-dup 1" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'], gtag,
conf_dict['options']['peakqval'])
else:
macs3cmd = "macs3 callpeak -t %s -n %s -f BED -g %s -q %s --keep-dup 1 --nomodel --extsize 100" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'], gtag,
conf_dict['options']['peakqval'])
wlog("peak calling with macs3: %s" % macs3cmd, logfile)
peaklog = sp(macs3cmd)
### ext peak from summit
wlog(
"extend peak summits to +/- %sbp" % conf_dict['options']['extend'],
logfile)
if not os.path.isfile(conf_dict['General']['outname'] +
"_summits.bed"):
ewlog(
"no macs3 results detected, check whether macs3 was correctly installed.",
logfile)
conf_dict['QC']['peaknumTotal'] = extsummit(
conf_dict['General']['outname'] + "_summits.bed",
conf_dict['results']['peakfile'],
int(conf_dict['options']['extend']))
if conf_dict['QC']['peaknumTotal'] < 1000:
ewlog(
"obtain < 1000 (%s) peaks, SELMA terminated" %
conf_dict['QC']['peaknumTotal'], logfile)
else:
wlog("obtain %s peaks" % conf_dict['QC']['peaknumTotal'], logfile)
return conf_dict
def step4_BULKcleavageBias(conf_dict, logfile):
### preparing mapping state dict
wlog('split fragments to strand specific cleavage sites', logfile)
if conf_dict['General']['format'] == "PE":
cmdplus = """awk '{OFS="\\t";print $1,$2,$2+1,".",".","+"}' %s > %s""" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'] + "_cleavage_plus.bed")
cmdminus = """awk '{OFS="\\t";print $1,$3-1,$3,".",".","-"}' %s > %s""" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'] + "_cleavage_minus.bed")
else:
cmdplus = """awk '{if($6=="+") print $0}' %s > %s""" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'] + "_cleavage_plus.bed")
cmdminus = """awk '{if($6=="-") print $0}' %s > %s""" % (
conf_dict['General']['outname'] + "_chromatin.bed",
conf_dict['General']['outname'] + "_cleavage_minus.bed")
tmplog = sp(cmdplus)
tmplog = sp(cmdminus)
wlog('pile up cleavage sites', logfile)
pluslog1 = sp("macs3 pileup -i %s -f BED --extsize 1 -o %s " %
(conf_dict['General']['outname'] + "_cleavage_plus.bed",
conf_dict['General']['outname'] + "_cleavage_plus.bdg"))
pluslog2 = sp(
"sort -k1,1 -k2,2n %s > %s" %
(conf_dict['General']['outname'] + "_cleavage_plus.bdg",
conf_dict['General']['outname'] + "_cleavage_plus_sorted.bdg"))
pluslog3 = sp("%s %s %s %s" %
(conf_dict['General']['bedGraphToBigWig'],
conf_dict['General']['outname'] +
"_cleavage_plus_sorted.bdg", conf_dict['options']['csize'],
conf_dict['General']['outname'] + "_cleavage_plus.bw"))
minuslog1 = sp("macs3 pileup -i %s -f BED --extsize 1 -o %s " %
(conf_dict['General']['outname'] + "_cleavage_minus.bed",
conf_dict['General']['outname'] + "_cleavage_minus.bdg"))
minuslog2 = sp(
"sort -k1,1 -k2,2n %s > %s" %
(conf_dict['General']['outname'] + "_cleavage_minus.bdg",
conf_dict['General']['outname'] + "_cleavage_minus_sorted.bdg"))
minuslog3 = sp(
"%s %s %s %s" %
(conf_dict['General']['bedGraphToBigWig'],
conf_dict['General']['outname'] + "_cleavage_minus_sorted.bdg",
conf_dict['options']['csize'],
conf_dict['General']['outname'] + "_cleavage_minus.bw"))
wlog("remove redundant position from the extended peak file", logfile)
cmduni = """sort -k 1,1 -k 2,2g -k 3,3g %s | %s merge -i - > %s""" % (
conf_dict['results']['peakfile'], conf_dict['General']['bedtools'],
conf_dict['General']['outname'] + "_summitEXTmerge.bed")
tmplog = sp(cmduni)
wlog('readin sequence from 2bit', logfile)
seq_dict = {}
inf = open(conf_dict['options']['csize'])
for line in inf:
chrm = line.split()[0]
seq_dict[chrm] = fetchseq_2bit_chrom(
conf_dict['General']['twoBitToFa'],
conf_dict['General']['sequence'], chrm)
inf.close()
conf_dict['results']['seqdict'] = seq_dict
wlog('calculate bias expected cleavages', logfile)
if conf_dict['General']['datatype'] == "DNase":
tmplog = bias_exp_cleavage_DNase(
conf_dict['General']['outname'],
conf_dict['General']['outname'] + "_summitEXTmerge.bed",
conf_dict['results']['biasMat'], conf_dict['options']['kmer'],
conf_dict['General']['bigWigSummary'],
conf_dict['General']['bedGraphToBigWig'],
conf_dict['results']['seqdict'])
else:
tmplog = bias_exp_cleavage_ATAC(
conf_dict['General']['outname'],
conf_dict['General']['outname'] + "_summitEXTmerge.bed",
conf_dict['results']['biasMat'], conf_dict['options']['kmer'],
conf_dict['General']['bigWigSummary'],
conf_dict['General']['bedGraphToBigWig'],
conf_dict['results']['seqdict'])
#
pluslog = sp(
"sort -k1,1 -k2,2n %s > %s" %
(conf_dict['General']['outname'] + "_biasExpCuts_plus.bdg",
conf_dict['General']['outname'] + "_biasExpCuts_plus_sorted.bdg"))
pluslog = sp(
"%s %s %s %s" %
(conf_dict['General']['bedGraphToBigWig'],
conf_dict['General']['outname'] + "_biasExpCuts_plus_sorted.bdg",
conf_dict['options']['csize'],
conf_dict['General']['outname'] + "_biasExpCuts_plus.bw"))
#
minuslog = sp(
"sort -k1,1 -k2,2n %s > %s" %
(conf_dict['General']['outname'] + "_biasExpCuts_minus.bdg",
conf_dict['General']['outname'] + "_biasExpCuts_minus_sorted.bdg"))
minuslog = sp(
"%s %s %s %s" %
(conf_dict['General']['bedGraphToBigWig'],
conf_dict['General']['outname'] + "_biasExpCuts_minus_sorted.bdg",
conf_dict['options']['csize'],
conf_dict['General']['outname'] + "_biasExpCuts_minus.bw"))
return conf_dict
def step5_SCcellClustering(conf_dict,logfile):
wlog('single-cell clustering analysis',logfile)
if conf_dict['options']['clustermethod'] == "PCAkm":
conf_dict['General']['scPackage'] = scClustering_PCAkm(conf_dict['General']['outname'],
conf_dict['options']['lowbiaspeak'],
conf_dict['options']['clusterNum'],
conf_dict['options']['topDim'],
int(conf_dict['options']['UMAP']))
if conf_dict['General']['scPackage'] == "noPackage":
wlog("umap was not installed, UMAP scatter plot will not be generated",logfile)
elif conf_dict['options']['clustermethod'] == "Seurat":
conf_dict['General']['scPackage'] = scClustering_Seurat(conf_dict['General']['outname'],
conf_dict['options']['lowbiaspeak'],
conf_dict['options']['topDim'],
int(conf_dict['options']['UMAP']))
if conf_dict['General']['scPackage'] == "noPackage":
wlog("Seurat related packages were not installed, skip single-cell clustering step",logfile)
elif conf_dict['options']['clustermethod'] == "ArchR":
conf_dict['General']['scPackage'] = scClustering_ArchR(conf_dict['General']['outname'],
conf_dict['General']['genome'],
conf_dict['options']['lowbiaspeak'],
conf_dict['options']['topDim'],
int(conf_dict['options']['UMAP']))
if conf_dict['General']['scPackage'] == "noPackage":
wlog("ArchR related packages were not installed, skip single-cell clustering step",logfile)
elif conf_dict['options']['clustermethod'] == "APEC":
conf_dict['General']['scPackage'] = scClustering_APEC(conf_dict['General']['outname'],
conf_dict['options']['lowbiaspeak'],
int(conf_dict['options']['UMAP']))
if conf_dict['General']['scPackage'] == "noPackage":
wlog("APEC related packages were not installed, skip single-cell clustering step",logfile)
elif conf_dict['options']['clustermethod'] == "Cicero":
conf_dict['General']['scPackage'] = scClustering_Cicero(conf_dict['General']['outname'],
conf_dict['options']['lowbiaspeak'],
int(conf_dict['options']['UMAP']))
if conf_dict['General']['scPackage'] == "noPackage":
wlog("Cicero related packages were not installed, skip single-cell clustering step",logfile)
if os.path.isfile("%s_scClusters.txt"%(conf_dict['General']['outname'])):
tmplist = []
inf = open("%s_scClusters.txt"%(conf_dict['General']['outname']))
for line in inf:
ll = line.strip().split("\t")
if ll[1] != "cluster":
tmplist.append(ll[1])
inf.close()
conf_dict['QC']['scClusters'] = len(set(tmplist))
else:
conf_dict['QC']['scClusters'] = 0
return conf_dict
def stepFinal_summary(conf_dict, logfile):
wlog('Collect results', logfile)
summarydir = 'summary/'
createDIR(summarydir)
if "biasfile" in conf_dict['results'] and os.path.isfile(
conf_dict['results']['biasfile']):
sp("mv %s %s" % (conf_dict['results']['biasfile'], summarydir))
sp("mv %s_summitEXT.bed %s" %
(conf_dict['General']['outname'], summarydir))
conf_dict['results']['umap'] = "NA"
if conf_dict['General']['mode'] == "bulk":
sp("mv %s_cleavage_plus.bw %s" %
(conf_dict['General']['outname'], summarydir))
sp("mv %s_cleavage_minus.bw %s" %
(conf_dict['General']['outname'], summarydir))
sp("mv %s_biasExpCuts_plus.bw %s" %
(conf_dict['General']['outname'], summarydir))
sp("mv %s_biasExpCuts_minus.bw %s" %
(conf_dict['General']['outname'], summarydir))
else:
sp("gzip %s_peakXcellMat.txt" % (conf_dict['General']['outname']))
sp("gzip %s_peakFeatures.txt" % (conf_dict['General']['outname']))
sp("mv %s_peakXcellMat.txt.gz %s" %
(conf_dict['General']['outname'], summarydir))
sp("mv %s_peakFeatures.txt.gz %s" %
(conf_dict['General']['outname'], summarydir))
sp("mv %s_scClusters.txt %s" %
(conf_dict['General']['outname'], summarydir))
if os.path.isfile("%s_clusteringUMAP.pdf" %
(conf_dict['General']['outname'])):
if conf_dict['options']['clustermethod'] in ["APEC", "Cicero"]:
sp("mv %s_clusteringUMAP.pdf %s/%s_clusteringTSNE.pdf" %
(conf_dict['General']['outname'], summarydir,
conf_dict['General']['outname']))
conf_dict['results']['umap'] = "%s_clusteringTSNE.pdf" % (
conf_dict['General']['outname'])
else:
sp("mv %s_clusteringUMAP.pdf %s" %
(conf_dict['General']['outname'], summarydir))
conf_dict['results']['umap'] = "%s_clusteringUMAP.pdf" % (
conf_dict['General']['outname'])
tmpresult = 'tmpResults/'
createDIR(tmpresult)
sp("mv %s_chromatin.bed %s" % (conf_dict['General']['outname'], tmpresult))
sp("mv %s_chrM.bed %s" % (conf_dict['General']['outname'], tmpresult))
sp("mv %s_summits.bed %s" % (conf_dict['General']['outname'], tmpresult))
sp("mv %s_peaks.xls %s" % (conf_dict['General']['outname'], tmpresult))
sp("mv %s_peaks.narrowPeak %s" %
(conf_dict['General']['outname'], tmpresult))
if conf_dict['General']['mode'] == "bulk":
sp("mv %s_cleavage_plus.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_cleavage_minus.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_cleavage_plus.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_cleavage_minus.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_cleavage_plus_sorted.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_cleavage_minus_sorted.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_biasExpCuts_plus.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_biasExpCuts_minus.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_biasExpCuts_plus_sorted.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_biasExpCuts_minus_sorted.bdg %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_summitEXTmerge.bed %s" %
(conf_dict['General']['outname'], tmpresult))
else:
sp("mv %s_highQcellReads.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_tmpSCreads.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_tmpSCpeaks.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_scOVcleavage.bed %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_scRscript.r %s" %
(conf_dict['General']['outname'], tmpresult))
sp("mv %s_highQcellReads.bed %s" %
(conf_dict['General']['outname'], tmpresult))
if conf_dict['options']['clustermethod'] == "ArchR":
sp("mv %s_ArchR %s" % (conf_dict['General']['outname'], tmpresult))
if conf_dict['options']['clustermethod'] == "APEC":
sp("mv %s_APEC %s" % (conf_dict['General']['outname'], tmpresult))
if conf_dict['options']['clustermethod'] == "Cicero":
sp("mv %s_Cicero %s" %
(conf_dict['General']['outname'], tmpresult))
if conf_dict['options']['keeptmp']:
wlog('--keeptmp not setting, keep intermediate results', logfile)
pass
else:
wlog('--keeptmp was not setting, remove intermediate results', logfile)
sp("rm -r tmpResults/")
wlog('Generate summary reports', logfile)
outf = open("%s_summaryReports.txt" % conf_dict['General']['outname'], 'w')
outf.write("#settings\n")
outf.write("mode\t%s\n" % (conf_dict['General']['mode']))
outf.write("fragments\t%s\n" % (conf_dict['General']['fragments']))
outf.write("data format\t%s\n" % (conf_dict['General']['format']))
outf.write("data type\t%s\n" % (conf_dict['General']['datatype']))
outf.write("genome version\t%s\n" % (conf_dict['General']['genome']))
outf.write("output name\t%s\n" % (conf_dict['General']['outname']))
outf.write("\n#parameters\n")
outf.write("peak extend size\t%s\n" % (conf_dict['options']['extend']))
outf.write("peak qvalue\t%s\n" % (conf_dict['options']['peakqval']))
outf.write("bias source\t%s\n" % (conf_dict['options']['bias']))
outf.write("k-mer\t%s\n" % (conf_dict['options']['kmer']))
if conf_dict['General']['mode'] == "sc":
outf.write("[sc]reads cutoff\t%s\n" %
(conf_dict['options']['readcutoff']))
outf.write("[sc]%low biaspeak\t" +
str(conf_dict['options']['lowbiaspeak']) + "\n")
outf.write("[sc]peak min reads\t%s\n" %
(conf_dict['options']['peakminreads']))
outf.write("[sc]peak max reads\t%s\n" %
(conf_dict['options']['peakmaxreads']))
outf.write("[sc]topN dimensions\t%s\n" %
(conf_dict['options']['topDim']))
outf.write("[sc]clustering method\t%s\n" %
(conf_dict['options']['clustermethod']))
# if conf_dict['options']['clustermethod'] == "PCAkm":
# outf.write("[sc]cluster number\t%s\n"%(conf_dict['options']['clusterNum']))
outf.write("\n#QC\n")
outf.write(
"total reads\t%s\n" %
(conf_dict['QC']['chrM_reads'] + conf_dict['QC']['chromatin_reads']))
outf.write("chromatin reads\t%s\n" % (conf_dict['QC']['chromatin_reads']))
outf.write("mtDNA reads\t%s\n" % (conf_dict['QC']['chrM_reads']))
outf.write("total peaks\t%s\n" % (conf_dict['QC']['peaknumTotal']))
if conf_dict['General']['mode'] == "sc":
outf.write("total single-cells\t%s\n" %
(conf_dict['QC']['totalcellnum']))
outf.write("high quality single-cells\t%s\n" %
(conf_dict['QC']['highQcellnum']))
outf.write("single-cells for clustering\t%s\n" %
(conf_dict['QC']['finalusecellnum']))
outf.write("reads in single-cells for clustering\t%s\n" %
(conf_dict['QC']['finalreadnum']))
if conf_dict['QC']['scClusters'] > 0:
outf.write("cluster number\t%s\n" %
(conf_dict['QC']['scClusters']))
outf.write("\n#output results\n")
outf.write("peaks (accessible regions)\t%s_summitEXT.bed\n" %
(conf_dict['General']['outname']))
if conf_dict['General']['mode'] == "bulk":
outf.write("observed cleavage (+ strand)\t%s_cleavage_plus.bw\n" %
(conf_dict['General']['outname']))
outf.write("observed cleavage (- strand)\t%s_cleavage_minus.bw\n" %
(conf_dict['General']['outname']))
outf.write(
"bias expected cleavage (+ strand)\t%s_biasExpCuts_plus.bw\n" %
(conf_dict['General']['outname']))
outf.write(
"bias expected cleavage (- strand)\t%s_biasExpCuts_minus.bw\n" %
(conf_dict['General']['outname']))
else:
outf.write("peak bias features\t%s_peakFeatures.txt.gz\n" %
(conf_dict['General']['outname']))
outf.write("peakXcell count\t%s_peakXcellMat.txt.gz\n" %
(conf_dict['General']['outname']))
if os.path.isfile("%s%s_scClusters.txt" %
(summarydir, conf_dict['General']['outname'])):
outf.write("single-cell cluster\t%s_scClusters.txt\n" %
(conf_dict['General']['outname']))
if os.path.isfile(summarydir + conf_dict['results']['umap']):
if conf_dict['options']['clustermethod'] in ["APEC", "Cicero"]:
outf.write("sc-cluster t-SNE\t%s\n" %
(conf_dict['results']['umap']))
else:
outf.write("sc-cluster UMAP\t%s\n" %
(conf_dict['results']['umap']))
outf.close()
### check pdflatex
QCdoc = """\documentclass[11pt,a4paper]{article}
\\usepackage{tabularx}
\\usepackage[english]{babel}
\\usepackage{array}
\\usepackage{graphicx}
\\usepackage{color}
\DeclareGraphicsExtensions{.eps,.png,.pdf,.ps}
\\begin{document}
\\title{SELMA summary reports for: %s}
\\vspace{-1cm}
\maketitle
\\tableofcontents
\\newpage
\\newpage
\section{Summary description}
\\begin{quotation}
Table 1 describes the input files and settings.
\end{quotation}
\\begin{table}[h]
\\small
\caption{ settings }\label{bstable}
\\begin{tabularx}{\\textwidth}{ |X|l| }
""" % (strlatexformat(conf_dict['General']['outname']))
### table1 prepare parameter
QCdoc += """
\hline
parameter & value \\\\
\hline
mode & %s \\\\
\hline
fragment file & %s \\\\
\hline
data format & %s \\\\
\hline
data type & %s \\\\
\hline
genome version & %s \\\\
\hline
output name & %s \\\\
\hline
\end{tabularx}
\end{table}
""" % (strlatexformat(conf_dict['General']['mode']),
strlatexformat(conf_dict['General']['fragments'].split("/")[-1]),
strlatexformat(conf_dict['General']['format']),
strlatexformat(conf_dict['General']['datatype']),
strlatexformat(conf_dict['General']['genome']),
strlatexformat(conf_dict['General']['outname']))
QCdoc += """
\\newpage
\\newpage
\section{parameters and options}
\\begin{quotation}
Table 2 describes the parameters and options.
\end{quotation}
\\begin{table}[h]
\\small
\caption{parameters and options}\label{bstable}
\\begin{tabularx}{\\textwidth}{ |X|l| }
\hline
parameter & value \\\\
\hline
peak extend size & %s \\\\
\hline
peak qvalue & %s \\\\
\hline
bias source & %s \\\\
\hline
k-mer & %s \\\\
\hline
""" % (strlatexformat(conf_dict['options']['extend']),
strlatexformat(conf_dict['options']['peakqval']),
strlatexformat(conf_dict['options']['bias']),
strlatexformat(conf_dict['options']['kmer']))
if conf_dict['General']['mode'] == "sc":
QCdoc += """
[sc]reads cutoff & %s \\\\
\hline
[sc]lowbias peak percent & %s \\\\
\hline
[sc]peak min reads & %s \\\\
\hline
[sc]peak max reads & %s \\\\
\hline
[sc]topN dimensions & %s \\\\
\hline
[sc]cluster methods & %s \\\\
\hline
""" % (
strlatexformat(conf_dict['options']['readcutoff']),
strlatexformat(conf_dict['options']['lowbiaspeak']),
strlatexformat(conf_dict['options']['peakminreads']),
strlatexformat(conf_dict['options']['peakmaxreads']),
strlatexformat(conf_dict['options']['topDim']),
strlatexformat(conf_dict['options']['clustermethod']),
)
QCdoc += """
\end{tabularx}
\end{table}
"""
QCdoc += """
\\newpage
\\newpage
\section{data quality}
\\begin{quotation}
Table 3 describes data Quality.
\end{quotation}
\\begin{table}[h]
\\small
\caption{data quality}\label{bstable}
\\begin{tabularx}{\\textwidth}{ |X|l| }
\hline
parameter & value \\\\
\hline
total reads & %s \\\\
\hline
chromatin reads & %s \\\\
\hline
mtDNA reads & %s \\\\
\hline
total peaks & %s \\\\
\hline
""" % (strlatexformat(conf_dict['QC']['chrM_reads'] +
conf_dict['QC']['chromatin_reads']),
strlatexformat(conf_dict['QC']['chromatin_reads']),
strlatexformat(conf_dict['QC']['chrM_reads']),
strlatexformat(conf_dict['QC']['peaknumTotal']))
if conf_dict['General']['mode'] == "sc":
QCdoc += """
[sc]total single-cells(sc) & %s \\\\
\hline
[sc]high quality sc & %s \\\\
\hline
[sc]sc for clustering & %s \\\\
\hline
[sc]reads in sc for clustering & %s \\\\
\hline
""" % (strlatexformat(conf_dict['QC']['totalcellnum']),
strlatexformat(conf_dict['QC']['highQcellnum']),
strlatexformat(conf_dict['QC']['finalusecellnum']),
strlatexformat(conf_dict['QC']['finalreadnum']))
if conf_dict['QC']['scClusters'] > 0:
QCdoc += """[sc]number of cluster & %s \\\\
\hline
""" % (strlatexformat(conf_dict['QC']['scClusters']))
QCdoc += """
\end{tabularx}
\end{table}
"""
QCdoc += """
\\newpage
\\newpage
\section{output results}
\\begin{quotation}
Table 3 describes output results (in the summary/ folder).
\end{quotation}
\\begin{table}[h]
\\small
\caption{output results}\label{bstable}
\\begin{tabularx}{\\textwidth}{ |X|l| }
\hline
parameter & value \\\\
\hline
peaks (accessible regions) & %s \\\\
\hline
""" % (strlatexformat(conf_dict['General']['outname'] + "_summitEXT.bed"))
if conf_dict['General']['mode'] == "bulk":
QCdoc += """
observed cleavage(+) & %s \\\\
\hline
observed cleavage(-) & %s \\\\
\hline
bias expected cleavage(+) & %s \\\\
\hline
bias expected cleavage(-) & %s \\\\
\hline
""" % (strlatexformat(conf_dict['General']['outname'] + "_cleavage_plus.bw"),
strlatexformat(conf_dict['General']['outname'] + "_cleavage_minus.bw"),
strlatexformat(conf_dict['General']['outname'] +
"_biasExpCuts_plus.bw"),
strlatexformat(conf_dict['General']['outname'] +
"_biasExpCuts_minus.bw"))
else:
QCdoc += """
peak bias feature & %s \\\\
\hline
peakXcell count & %s \\\\
\hline
""" % (strlatexformat(conf_dict['General']['outname'] +
"_peakFeatures.txt.gz"),
strlatexformat(conf_dict['General']['outname'] +
"_peakXcellMat.txt.gz"))
if os.path.isfile("%s%s_scClusters.txt" %
(summarydir, conf_dict['General']['outname'])):
QCdoc += """single-cell cluster & %s \\\\
\hline
""" % (strlatexformat(conf_dict['General']['outname'] + "_scClusters.txt"))
if os.path.isfile(summarydir + conf_dict['results']['umap']):
if conf_dict['options']['clustermethod'] in ["APEC", "Cicero"]:
dimRedTerm = "t-SNE"
else:
dimRedTerm = "UMAP"
QCdoc += """sc-cluster %s & %s \\\\
\hline
""" % (dimRedTerm,
strlatexformat(conf_dict['General']['outname'] + "_scClusters.txt"))
QCdoc += """
\end{tabularx}
\end{table}
"""
if os.path.isfile(summarydir + conf_dict['results']['umap']):
if conf_dict['options']['clustermethod'] in ["APEC", "Cicero"]:
dimRedTerm = "t-SNE"
else:
dimRedTerm = "UMAP"
QCdoc += """
\\newpage
\\newpage
\section{%s scatter plot}
The 2-dim scatter plot represent the %s results. Each dot represents an individual cell and the color represents cluster labels
\\begin{figure}[h]
\caption{cross-validation curve for lambda decision} \label{fig:profileunion}
\setlength{\\abovecaptionskip}{0pt}
\setlength{\\belowcaptionskip}{10pt}
\centering
{\includegraphics[width=0.8\\textwidth]{%s}}
\end{figure}
""" % (dimRedTerm, dimRedTerm, summarydir + conf_dict['results']['umap'])
QCdoc += """
\end{document}
"""
latexfile = conf_dict['General']['outname'] + '_summaryReports.tex'
outf = open(latexfile, 'w')
outf.write(QCdoc)
outf.close()
check_latex = sp('which pdflatex')
if check_latex[0].decode("ascii") == "":
wlog(
'pdflatex was not installed, SELMA will not generate pdf version of summary report. Please copy the %s to an environment with pdflatex installed and complie the pdf file'
% (conf_dict['General']['outname'] + '_summaryReports.tex'),
logfile)
else:
cmd = "pdflatex %s" % (latexfile)
tmpobj = sp(cmd)
tmpobj = sp(cmd)
#tmpobj = sp(cmd2)
tmpobj = sp("rm %s_summaryReports.aux" %
conf_dict['General']['outname'])
tmpobj = sp("rm %s_summaryReports.log" %
conf_dict['General']['outname'])
tmpobj = sp("rm %s_summaryReports.toc" %
conf_dict['General']['outname'])
return conf_dict