Mira assembly -shell script

[robelb4]

hello,

I'm a beginner in Assembly and shell scripting. My question is I already have pair-end reads generated using solexa. and on the MIRA documentation about mira input file processing (processing to make /1/2 illumina naming scheme), there is a shell script called prepdata.sh .

but my two reads are not form NCBI website;they're already unziped, so how can i modify the script to just work for an already unzipped file???

thanks,
R

[gaffa]

I'm not familar with this, but is this the script in question? (from http://mira-assembler.sourceforge.ne...deToMIRA.html):

######################################################################
#######
####### Prepare paired-end Solexa downloaded from NCBI
#######
######################################################################

# srrname:    is the SRR name as downloaded form NCBI SRA
# numreads:   maximum number of forward (and reverse) reads to take from
#              each file. Just to avoid bacterial projects with a coverage
#              of 200 or so.
# strainname: name of the strain which was re-sequenced

srrname="SRR030257"
numreads=5000000
strainname="REL8593A"

################################

numlines=$((4*${numreads}))

# put "/1" Solexa reads into file
echo "Copying ${numreads} reads from _1 (forward reads)"
zcat ../origdata/${srrname}_1.fastq.gz | head -${numlines} | sed -e 's/SRR[0-9.]*/&\/1/' >${strainname}-${numreads}_in.solexa.fastq

# put "/2" Solexa reads into file
echo "Copying ${numreads} reads from _2 (reverse reads)"
zcat ../origdata/${srrname}_2.fastq.gz | head -${numlines} | sed -e 's/SRR[0-9.]*/&\/2/' >>${strainname}-${numreads}_in.solexa.fastq

# make file with strainnames
echo "Creating file with strain names for copied reads (this may take a while)."
grep "@SRR" ${strainname}-${numreads}_in.solexa.fastq | cut -f 1 -d ' ' | sed -e 's/@//' -e "s/$/ ${strainname}/" >>${strainname}-${numreads}_straindata_in.txt

If so, then it should probably just be a matter of changing the "zcat" commands to "cat" (aswell as changing the appropriate file names of course, e.g. it shouldn't say ".gz"). I haven't tried this or anything but it should work. "zcat" is for printing gzipped files to screen, cat is for normal, non-zipped files.

[robelb4]

gaffa,

Thanks, that's what I thought and tried, but it's i'm getting a 'No such file ..' error . May be I misplaced the reads directory.

now I know the command is correct.

thanks
Robel