def __cir_mols(self, exon_level_MinDepth, min_depth=0):
l_brief_name = [
self['sam_info']['samp_brief'][samp] for samp in self['sample']
]
ltype = "circ_mols"
inFile = "%s/04.CIRC_info" % (self.statInfo)
outFile = "%s/04.CIRC_info" % (self.statInfo)
circ_Mols = m_cnt.CountInfo(inFile, l_brief_name, ltype, outFile)
circ_Mols.load_mat(exon_level_MinDepth, gen_col=2)
circ_Mols.sam_tot_reads()
M_gen_len = {}
for gene in circ_Mols.gene:
M_gen_len[gene] = {
'max_len': 200
} #### Please ONLY USE samples with less index, that sequenced by illumina X10 !!!!!!
circ_Mols.cal_RPKM(M_gen_len, self.tophat)
out_file_RPKM = "%s/merge.%s.RPKM.xls" % (outFile, ltype)
cirRPKM_mat = m_mat.Matrix_info(out_file_RPKM,
inf_column=1,
in_dtype="float")
cirRPKM_mat.load_mat()
l_cir_genes = cirRPKM_mat.colname
for brief_name in l_brief_name:
idx_rowname = cirRPKM_mat.rowname.index(brief_name)
self.l_cirRNA_FPKM[brief_name] = cirRPKM_mat.matrix[:, idx_rowname]
self.__get_cirRNA_MOLs(
) # get cirRNA mols using ERCC_FPKM, ERCC_MOLs and cirRNA_FPKM
def __load_Count(self): # for ERCC count Stat
l_breif_samp = [ self['sam_info']['samp_brief'][samp] for samp in self['sample'] ]
genome_gtf = self['infile']['anno_file']
Gtf_Info = m_gtf.GTFFeature( genome_gtf )
ERCC_info = m_cnt.CountInfo( self.HTS_k,l_breif_samp,"dexseq_ERCC_RGCPloyA",self.HTS )
ERCC_info.load_mat()
ERCC_info.cal_RPKM( Gtf_Info.gene,self.tophat )
def RPKM_novo_trans(self):
l_brief_samp = [ "%s" % ( self['samp_info']['samp_brief'][samp] ) for samp in self['samp'] ]
unknown_GTF = "%s/novo_lnc_raw.combined.gtf" % ( self.data_dir )
Gtf_Info = m_gtf.GTFFeature( unknown_GTF )
Gtf_Info.get_intergenic( self['infile']['intragenic_bed'] )
Cnt_Info = m_cnt.CountInfo( self.HTS_u, l_brief_samp, "dexseq_NeoRaw", self.HTS )
Cnt_Info.generate_mat()
Cnt_Info.load_mat()
Cnt_Info.cal_RPKM( Gtf_Info.gene,self.tophat )
rpkm_file = "%s/merge.%s.RPKM.xls" % ( self.HTS,"dexseq_NeoRaw" )
Gtf_Info.load_gene_RPKM( rpkm_file )
Gtf_Info.output_GTF()
Gtf_Info.get_gene_info()
def Basic_Stat(self):
"""
Stat for QC, Tophat mapping, ERCC RGC count
"""
out_file = "%s/01.BasicInfo_QC_map_SpikeIn.xls" % (self.statInfo)
f_out_file = open(out_file, "w")
out_info = "Sample\tBrief_samp\tRaw_Reads\tClean_Reads\t" +\
"Pre_Map_Reads\tAligned_Reads\tHTSseq_Known_Reads\t" +\
"HTSeq_Refseq_Reads\tHTSeq_NONCODE_V4_Reads\tHTSeq_Nsmb_Reads\t" +\
"RFP_Reads\tGFP_Reads\tCRE_Reads\tERCC_Reads\t" +\
"RFP_Mols\tGFP_Mols\tCRE_Mols\tERCC_Mols"
print >> f_out_file, out_info
l_breif_samp = [
self['sam_info']['samp_brief'][samp] for samp in self['sample']
]
'''
Load refseq reads
'''
HTS_info = m_cnt.CountInfo(self.HTS_k, l_breif_samp, "dexseq_clean",
self.HTS)
HTS_info.load_mat()
HTS_info.sam_tot_reads()
self.__get_HTS_clean_split()
Refseq_info = m_cnt.CountInfo(self.HTS_k, l_breif_samp,
"dexseq_clean_refseq", self.HTS)
Refseq_info.load_mat()
Refseq_info.sam_tot_reads()
NONCODE_info = m_cnt.CountInfo(self.HTS_k, l_breif_samp,
"dexseq_clean_NONCODE", self.HTS)
NONCODE_info.load_mat()
NONCODE_info.sam_tot_reads()
NSMB_info = m_cnt.CountInfo(self.HTS_k, l_breif_samp,
"dexseq_clean_NSMB", self.HTS)
NSMB_info.load_mat()
NSMB_info.sam_tot_reads()
'''
Load other information
'''
for idx, samp in enumerate(self['sample']):
brief_name = self['sam_info']['samp_brief'][samp]
QC_log = "%s/%s/log" % (self.cln, samp)
Tophat_log = "%s/%s/align_summary.txt" % (self.tophat, brief_name)
HTSeq_SpikeIn = "%s/%s/%s.dexseq_ERCC_RGCPloyA.txt" % (
self.HTS_k, brief_name, brief_name)
QcStat_info = Stat.QcStat(QC_log)
MapStat_info = Stat.TophatStat(Tophat_log)
SpikeIn_info = Stat.SpikeIn(HTSeq_SpikeIn,
self['infile']['ercc_info'])
QcStat_info.read_infile()
MapStat_info.read_infile()
SpikeIn_info.load_HTS_file()
pre_map_read = MapStat_info['statInfo']['totalRead']
aligned_read = MapStat_info['statInfo']['mappedRead']
if self['sam_info']['data_type'][samp] == "PE":
HTSseq_read = self.__get_HTS_reads(HTS_info, samp) * 2
Refseq_read = self.__get_HTS_reads(Refseq_info, samp) * 2
NONCODE_read = self.__get_HTS_reads(NONCODE_info, samp) * 2
NSMB_read = self.__get_HTS_reads(NSMB_info, samp) * 2
read_RFP = SpikeIn_info.RGC_count['RGC-mRFP'] * 2
read_GFP = SpikeIn_info.RGC_count['RGC-GFP'] * 2
read_CRE = SpikeIn_info.RGC_count['RGC-CRE'] * 2
read_ERCC = SpikeIn_info.ERCC_total * 2
else:
HTSseq_read = self.__get_HTS_reads(HTS_info, samp)
Refseq_read = self.__get_HTS_reads(Refseq_info, samp)
NONCODE_read = self.__get_HTS_reads(NONCODE_info, samp)
NSMB_read = self.__get_HTS_reads(NSMB_info, samp)
read_RFP = SpikeIn_info.RGC_count['RGC-mRFP']
read_GFP = SpikeIn_info.RGC_count['RGC-GFP']
read_CRE = SpikeIn_info.RGC_count['RGC-CRE']
read_ERCC = SpikeIn_info.ERCC_total
mol_RFP = self['sam_info']['RFP_mols'][samp]
mol_GFP = self['sam_info']['GFP_mols'][samp]
mol_CRE = self['sam_info']['CRE_mols'][samp]
mol_ERCC = self['sam_info']['dilute'][samp] * 6.023 * 10**10
out_info = "%s\t%s\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%1.2e\t%1.2e\t%1.2e\t%1.2e" \
% ( samp, brief_name, QcStat_info.raw_reads, QcStat_info.cln_reads, \
pre_map_read, aligned_read, HTSseq_read, \
Refseq_read,NONCODE_read,NSMB_read, \
read_RFP, read_GFP,read_CRE, read_ERCC, \
mol_RFP , mol_GFP ,mol_CRE , mol_ERCC )
print >> f_out_file, out_info
f_out_file.close()
def Basic_Stat(self):
"""
Stat for QC, Tophat mapping, ERCC RGC count
"""
out_file = "%s/01.BasicInfo_QC_map_SpikeIn.xls" % (self.statInfo)
f_out_file = open(out_file, "w")
out_info = "Sample\tBrief_samp\tRaw_Reads\tClean_Reads\t" +\
"Pre_Map_Reads\tAligned_Reads\tRefseq_Reads\t" +\
"Circular_genes\tCircular_exons\tCircular_reads\t"+\
"RFP_Reads\tGFP_Reads\tCRE_Reads\tERCC_Reads\t" +\
"RFP_Mols\tGFP_Mols\tCRE_Mols\tERCC_Mols"
print >> f_out_file, out_info
l_breif_samp = [
self['sam_info']['samp_brief'][samp] for samp in self['sample']
]
'''
Load refseq reads
'''
Refseq_info = m_cnt.CountInfo(self.HTS_k, l_breif_samp, "dexseq_clean",
self.HTS)
Refseq_info.load_mat()
Refseq_info.sam_tot_reads()
'''
Load circular reads. Run this step after CIRC_Stat.
'''
exon_level_MinDepth = "%s/04.CIRC_info/CIRC_PE_result.merge_exon_level.MinDepth_%d.xls" % (
self.statInfo, self.min_depth)
gene_level_MinDepth = "%s/04.CIRC_info/CIRC_PE_result.merge_gene_level.MinDepth_%d.xls" % (
self.statInfo, self.min_depth)
exon_level_mat = m_mat.Matrix_info(exon_level_MinDepth, 2)
gene_level_mat = m_mat.Matrix_info(gene_level_MinDepth, 1)
exon_level_mat.load_mat()
gene_level_mat.load_mat()
# How many genes have cirRNA with depth>2 in one junction of this gene for a given sample?
# How many exons have cirRNA with depth>2 in one junction for a given sample?
# Sum( reads ) for a given sample?
np_gene_cirs_samp = np.sum(gene_level_mat.matrix >= self.min_depth,
axis=0)
np_exon_cirs_samp = np.sum(exon_level_mat.matrix >= self.min_depth,
axis=0)
np_exon_read_samp = np.sum(exon_level_mat.matrix, axis=0)
'''
Load other information
'''
for idx, samp in enumerate(self['sample']):
brief_name = self['sam_info']['samp_brief'][samp]
QC_log = "%s/%s/log" % (self.cln, samp)
Tophat_log = "%s/%s/align_summary.txt" % (self.tophat, brief_name)
HTSeq_SpikeIn = "%s/%s/%s.dexseq_ERCC_RGCPloyA.txt" % (
self.HTS_k, brief_name, brief_name)
QcStat_info = Stat.QcStat(QC_log)
MapStat_info = Stat.TophatStat(Tophat_log)
SpikeIn_info = Stat.SpikeIn(HTSeq_SpikeIn,
self['infile']['ercc_info'])
QcStat_info.read_infile()
MapStat_info.read_infile()
SpikeIn_info.load_HTS_file()
pre_map_read = MapStat_info['statInfo']['totalRead']
aligned_read = MapStat_info['statInfo']['mappedRead']
refseq_read = self.__get_refseq_reads(Refseq_info, samp)
cir_genes = np_gene_cirs_samp[idx]
cir_exons = np_exon_cirs_samp[idx]
cir_reads = np_exon_read_samp[idx]
read_RFP = SpikeIn_info.RGC_count['RGC-mRFP']
read_GFP = SpikeIn_info.RGC_count['RGC-GFP']
read_CRE = SpikeIn_info.RGC_count['RGC-CRE']
read_ERCC = SpikeIn_info.ERCC_total
mol_RFP = self['sam_info']['RFP_mols'][samp]
mol_GFP = self['sam_info']['GFP_mols'][samp]
mol_CRE = self['sam_info']['CRE_mols'][samp]
mol_ERCC = self['sam_info']['dilute'][samp] * 6.023 * 10**10
out_info = "%s\t%s\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%d\t%1.2e\t%1.2e\t%1.2e\t%1.2e" \
% ( samp, brief_name, QcStat_info.raw_reads, QcStat_info.cln_reads, \
pre_map_read, aligned_read, refseq_read, \
cir_genes, cir_exons , cir_reads, \
read_RFP, read_GFP,read_CRE, read_ERCC, \
mol_RFP , mol_GFP ,mol_CRE , mol_ERCC )
print >> f_out_file, out_info
f_out_file.close()