Patch bug in the inverted repeats annotation
Realize the annotation on the normalized chromosome Former-commit-id: 73e4e016d7bfe23fc92e4071efb0a4629eb8e21f Former-commit-id: d44249c0d078d475ca80386bb428fb660289b0dd
This commit is contained in:
@ -100,7 +100,7 @@ pushTmpDir ORG.normalize
|
|||||||
tmpLSC="tmp_$$_LSC.fasta"
|
tmpLSC="tmp_$$_LSC.fasta"
|
||||||
tmpSSC="tmp_$$_SSC.fasta"
|
tmpSSC="tmp_$$_SSC.fasta"
|
||||||
|
|
||||||
# Extract the first SC present in between the two IRs
|
# Extract the central SC present in between the two IRs
|
||||||
# considering it as LSC
|
# considering it as LSC
|
||||||
|
|
||||||
let "beginLSC=$endIR1+1"
|
let "beginLSC=$endIR1+1"
|
||||||
@ -110,7 +110,7 @@ pushTmpDir ORG.normalize
|
|||||||
strandLSC="${IR[1]}"
|
strandLSC="${IR[1]}"
|
||||||
|
|
||||||
|
|
||||||
# Extract the second SC present in two parts
|
# Extract the external SC present in two parts
|
||||||
# Considering it as SSC
|
# Considering it as SSC
|
||||||
|
|
||||||
let "beginSSC=$endIR2+1"
|
let "beginSSC=$endIR2+1"
|
||||||
@ -130,16 +130,17 @@ pushTmpDir ORG.normalize
|
|||||||
|
|
||||||
# Actually this is the oposite LSC is SSC and SSC is LSC
|
# Actually this is the oposite LSC is SSC and SSC is LSC
|
||||||
|
|
||||||
# Exchange the SSC and LSC sequences
|
# Exchanges the SSC and LSC sequences
|
||||||
mv ${tmpSSC} ${tmpfasta1}
|
mv ${tmpSSC} ${tmpfasta1}
|
||||||
mv ${tmpLSC} ${tmpSSC}
|
mv ${tmpLSC} ${tmpSSC}
|
||||||
mv ${tmpfasta1} ${tmpLSC}
|
mv ${tmpfasta1} ${tmpLSC}
|
||||||
|
|
||||||
# Exchange the IRa and IRb sequences
|
# Exchanges the IRa and IRb sequences
|
||||||
mv ${tmpIR1} ${tmpfasta1}
|
mv ${tmpIR1} ${tmpfasta1}
|
||||||
mv ${tmpIR2} ${tmpIR1}
|
mv ${tmpIR2} ${tmpIR1}
|
||||||
mv ${tmpfasta1} ${tmpIR2}
|
mv ${tmpfasta1} ${tmpIR2}
|
||||||
|
|
||||||
|
# Exchanges the strand of both the Single copies
|
||||||
tmp=${strandSSC}
|
tmp=${strandSSC}
|
||||||
strandSSC=${strandLSC}
|
strandSSC=${strandLSC}
|
||||||
strandLSC=${tmp}
|
strandLSC=${tmp}
|
||||||
@ -161,7 +162,7 @@ pushTmpDir ORG.normalize
|
|||||||
# Merges the four parts of the genome.
|
# Merges the four parts of the genome.
|
||||||
cat ${tmpLSC} ${tmpIR2} ${tmpSSC} ${tmpIR1} | joinfasta
|
cat ${tmpLSC} ${tmpIR2} ${tmpSSC} ${tmpIR1} | joinfasta
|
||||||
|
|
||||||
|
exit 1
|
||||||
|
|
||||||
popTmpDir
|
popTmpDir
|
||||||
|
|
||||||
|
@ -12,6 +12,15 @@ function lookForIR {
|
|||||||
|
|
||||||
local REPEATS="${MATCHES/.*/}.repseek"
|
local REPEATS="${MATCHES/.*/}.repseek"
|
||||||
|
|
||||||
|
# Blast columns:
|
||||||
|
# query id, subject id, % identity, alignment length, mismatches, gap opens, q. start, q. end, s. start, s. end, evalue, bit score
|
||||||
|
# We keep blast matches if :
|
||||||
|
# The match is longer than 1000
|
||||||
|
# The identity is higher than 80%
|
||||||
|
#
|
||||||
|
# The match file has the following format:
|
||||||
|
# LSC/SSC begin end same_strand=1/diff_strand=0
|
||||||
|
|
||||||
loginfo "Locating SSC and LSC by similarity..."
|
loginfo "Locating SSC and LSC by similarity..."
|
||||||
blastn -db ${SCDB} \
|
blastn -db ${SCDB} \
|
||||||
-query ${QUERY} \
|
-query ${QUERY} \
|
||||||
|
@ -8,6 +8,12 @@ repeats = open(sys.argv[2])
|
|||||||
chloro = {'LSC' : [], 'SSC' : [] }
|
chloro = {'LSC' : [], 'SSC' : [] }
|
||||||
chlorosize =0
|
chlorosize =0
|
||||||
|
|
||||||
|
# We scan the blast matches:
|
||||||
|
# We build a vector with one position per base pair counting the matches
|
||||||
|
|
||||||
|
# The match file has the following format:
|
||||||
|
# LSC/SSC begin end same_strand=1/diff_strand=0
|
||||||
|
|
||||||
for line in data:
|
for line in data:
|
||||||
parts = line.strip().split()
|
parts = line.strip().split()
|
||||||
if len(parts) >= 4:
|
if len(parts) >= 4:
|
||||||
@ -16,7 +22,8 @@ for line in data:
|
|||||||
end = int(parts[2])
|
end = int(parts[2])
|
||||||
direction = int(parts[3])
|
direction = int(parts[3])
|
||||||
|
|
||||||
|
# Change the code of the direction:
|
||||||
|
# reverse complement = -1
|
||||||
if direction==0:
|
if direction==0:
|
||||||
direction=-1
|
direction=-1
|
||||||
|
|
||||||
@ -39,32 +46,49 @@ maxLSC = float(max(abs(n) for n in chloro['LSC']))
|
|||||||
chloro['SSC']=[n / maxSSC for n in chloro['SSC']]
|
chloro['SSC']=[n / maxSSC for n in chloro['SSC']]
|
||||||
chloro['LSC']=[n / maxLSC for n in chloro['LSC']]
|
chloro['LSC']=[n / maxLSC for n in chloro['LSC']]
|
||||||
|
|
||||||
|
|
||||||
scoreMax=0
|
scoreMax=0
|
||||||
|
len1Max=0
|
||||||
|
len2Max=0
|
||||||
|
|
||||||
imax = len(chloro['LSC'])
|
imax = len(chloro['LSC'])
|
||||||
|
|
||||||
for line in repeats:
|
for line in repeats:
|
||||||
parts = line.strip().split()
|
parts = line.strip().split()
|
||||||
|
|
||||||
|
# First repeat position and length
|
||||||
|
# (position start at 0)
|
||||||
pos1 = int(parts[1]) -1
|
pos1 = int(parts[1]) -1
|
||||||
len1 = int(parts[3])
|
len1 = int(parts[3])
|
||||||
|
|
||||||
|
# Second repeat position and length
|
||||||
|
# (position start at 0)
|
||||||
pos2 = int(parts[2]) -1
|
pos2 = int(parts[2]) -1
|
||||||
len2 = int(parts[4])
|
len2 = int(parts[4])
|
||||||
|
|
||||||
|
# Location of the central single copy
|
||||||
|
# - in between the two IR -
|
||||||
c_begin = min(pos1 + len1,imax)
|
c_begin = min(pos1 + len1,imax)
|
||||||
c_end = min(pos2,imax)
|
c_end = min(pos2,imax)
|
||||||
|
|
||||||
|
# Location of the external single copy
|
||||||
|
# - in between the two IR -
|
||||||
o_max = min(pos1 ,imax)
|
o_max = min(pos1 ,imax)
|
||||||
o_min = min(pos2 + len2, imax)
|
o_min = min(pos2 + len2, imax)
|
||||||
|
|
||||||
c_lsc = sum(abs(chloro['LSC'][n]) for n in range(c_begin,c_end))
|
# count of coherent matches for LSC and SSC on the central single copy
|
||||||
c_ssc = sum(abs(chloro['SSC'][n]) for n in range(c_begin,c_end))
|
c_lsc = abs(sum(chloro['LSC'][n] for n in range(c_begin,c_end)))
|
||||||
|
c_ssc = abs(sum(chloro['SSC'][n] for n in range(c_begin,c_end)))
|
||||||
|
|
||||||
o_lsc = sum(abs(chloro['LSC'][n]) for n in range(0,o_max))
|
# count of coherent matches for LSC and SSC on the external single copy
|
||||||
o_ssc = sum(abs(chloro['SSC'][n]) for n in range(0,o_max))
|
# this score is in two parts before the first copy and after the second
|
||||||
|
o_lsc = sum(chloro['LSC'][n] for n in range(0,o_max))
|
||||||
|
o_ssc = sum(chloro['SSC'][n] for n in range(0,o_max))
|
||||||
|
|
||||||
o_lsc += sum(abs(chloro['LSC'][n]) for n in range(o_min,len(chloro['LSC'])))
|
o_lsc += sum(chloro['LSC'][n] for n in range(o_min,imax))
|
||||||
o_ssc += sum(abs(chloro['SSC'][n]) for n in range(o_min,len(chloro['SSC'])))
|
o_ssc += sum(chloro['SSC'][n] for n in range(o_min,imax))
|
||||||
|
|
||||||
|
o_lsc = abs(o_lsc)
|
||||||
|
o_ssc = abs(o_ssc)
|
||||||
|
|
||||||
c = float(c_lsc + c_ssc)
|
c = float(c_lsc + c_ssc)
|
||||||
o = float(o_lsc + o_ssc)
|
o = float(o_lsc + o_ssc)
|
||||||
@ -78,10 +102,9 @@ for line in repeats:
|
|||||||
o_ssc /= o
|
o_ssc /= o
|
||||||
|
|
||||||
score = ((c_lsc - c_ssc) ** 2 + (o_lsc - o_ssc) ** 2) / 2.0
|
score = ((c_lsc - c_ssc) ** 2 + (o_lsc - o_ssc) ** 2) / 2.0
|
||||||
|
|
||||||
# print >>sys.stderr,"c.lsc = %f c.ssc = %f o.lsc = %f o.ssc = %f score = %6.4f (len=%d)" % (c_lsc,c_ssc,o_lsc,o_ssc,score,len1)
|
# print >>sys.stderr,"c.lsc = %f c.ssc = %f o.lsc = %f o.ssc = %f score = %6.4f (len=%d)" % (c_lsc,c_ssc,o_lsc,o_ssc,score,len1)
|
||||||
|
|
||||||
if (score > scoreMax):
|
if (score >= scoreMax) and ((len1 > len1Max) or (len2 > len2Max)):
|
||||||
scoreMax = score
|
scoreMax = score
|
||||||
pos1Max = pos1
|
pos1Max = pos1
|
||||||
pos2Max = pos2
|
pos2Max = pos2
|
||||||
@ -99,8 +122,8 @@ c_ssc = sum(chloro['SSC'][n] for n in range(c_begin,c_end))
|
|||||||
o_lsc = sum(chloro['LSC'][n] for n in range(0,o_max))
|
o_lsc = sum(chloro['LSC'][n] for n in range(0,o_max))
|
||||||
o_ssc = sum(chloro['SSC'][n] for n in range(0,o_max))
|
o_ssc = sum(chloro['SSC'][n] for n in range(0,o_max))
|
||||||
|
|
||||||
o_lsc += sum(chloro['LSC'][n] for n in range(o_min,len(chloro['LSC'])))
|
o_lsc += sum(chloro['LSC'][n] for n in range(o_min,imax))
|
||||||
o_ssc += sum(chloro['SSC'][n] for n in range(o_min,len(chloro['SSC'])))
|
o_ssc += sum(chloro['SSC'][n] for n in range(o_min,imax))
|
||||||
|
|
||||||
if abs(c_lsc) > abs(c_ssc):
|
if abs(c_lsc) > abs(c_ssc):
|
||||||
center = "LSC"
|
center = "LSC"
|
||||||
|
@ -34,7 +34,7 @@ pushTmpDir ORG.organnot
|
|||||||
loginfo "Done."
|
loginfo "Done."
|
||||||
|
|
||||||
loginfo "Annotating the Inverted repeats and Single copies (LSC and SSC)..."
|
loginfo "Annotating the Inverted repeats and Single copies (LSC and SSC)..."
|
||||||
${PROG_DIR}/detectors/ir/bin/go_ir.sh ${QUERY} > "${RESULTS}.annot"
|
${PROG_DIR}/detectors/ir/bin/go_ir.sh "${RESULTS}.norm.fasta" > "${RESULTS}.annot"
|
||||||
loginfo "Done."
|
loginfo "Done."
|
||||||
|
|
||||||
loginfo "Annotating the tRNA..."
|
loginfo "Annotating the tRNA..."
|
||||||
|
Reference in New Issue
Block a user