Talk:Perl - Revision history

Christoph at 08:46, 8 September 2007

2007-09-08T08:46:11Z

Show changes

Christoph at 08:40, 8 September 2007

2007-09-08T08:40:48Z

Show changes

Christoph at 04:58, 1 July 2007

2007-07-01T04:58:27Z

Show changes

Christoph at 23:48, 4 June 2007

2007-06-04T23:48:23Z

Christoph at 05:17, 29 May 2007

2007-05-29T05:17:08Z

Christoph at 00:46, 27 April 2007

2007-04-27T00:46:19Z

Christoph at 00:01, 27 April 2007

2007-04-27T00:01:57Z

Christoph at 23:51, 13 April 2007

2007-04-13T23:51:16Z

New page

==Perl and [[MySQL]]==
This section will just list a bunch of random examples. They are useful to give you an idea of what you can do with Perl, MySQL, and the [[:Category:Linux Command Line Tools|CLI]].

''Note: These examples were taken from my course in Comparative Microbial Genomics at CBS (in Denmark).''

<pre>
mysql -B -e "update cmp_genomics.features set note = '' where user = USER() and note not like 'tcs%' or note is null"
foreach accession (AE017042 AE016879 AL111168 AL645882 AP008232 AP009048 BA000021 CP000034)
foreach type (ecf s54 s70)
cat source/$accession.proteins.$type.sigmas.hmmsearch | \
perl -ne 'next unless /^CDS_(\d+)\-(\d+)_DIR([\-\+]+)\s+([0-9\-\.e]+)\s+([0-9\-\.e]+)\s+/;\
my ($start,$stop,$dir,$score,$evalue) = ($1,$2,$3,$4,$5);\
next unless $score > 0;\
print "update cmp_genomics.features set note = \"Sigma Factor '$type'\" \
where start=$start and stop = $stop and user = user() and accession = \"'$accession'\";\n";'\
| mysql
end
end
</pre>
<pre>
mysql -B -e "update cmp_genomics.features set note = '' where user = USER() and note not like 'sigma%' or note is null"
foreach accession (AE017042 AE016879 AL111168 AL645882 AP008232 AP009048 BA000021 CP000034)
foreach type (RRreciever HisKA_1 HisKA_2 HisKA_3 HWE_HK)
cat source/$accession.$type.TCS.hmmsearch | \
perl -ne 'next unless /^CDS_(\d+)\-(\d+)_DIR([\-\+]+)\s+([0-9\-\.e]+)\s+([0-9\-\.e]+)\s+/;\
my ($start,$stop,$dir,$score,$evalue) = ($1,$2,$3,$4,$5);\
next unless $score > 0; \
print "update cmp_genomics.features set note = \"TCS '$type'\" \
where start=$start and stop = $stop and user = user() and accession = \"'$accession'\";\n";'\
| mysql -B
end
end
</pre>
<pre>
rm -f source/all.16s.fsa
foreach accession (AE017042 AE016879 AL111168 AL645882 AP008232 AP009048 BA000021 CP000034)
mysql -B -N -D cmp_genomics -e "select substr(seq,start,stop-start+1) as sequence,\
replace(substring_index(genomes.organism,' ',4),' ','_') as name,dir from genomes,features where \
genomes.accession = features.accession and genomes.user = features.user and \
features.user = user() and featuretype='rRNA' and features.accession = '$accession'" | \
perl -ane '$F[0] = reverse ( $F[0] ) if $F[2] eq "-";$F[0] =~ tr/ATGC/TACG/ if $F[2] eq "-";\
print ">$F[1]\n"; \
for ( $x = 0 ; $x < length ( $F[0] ) ; $x = $x+60) { print substr ( $F[0] , $x , 60),"\n"; } last;' \
>> source/all.16s.fsa
end
less source/all.16s.fsa
</pre>
<pre>
cat ~/Ex3/source/CP000034.codon-usage.aa.list \
| perl -ane 'next unless $F[1] =~ /^[A-Z]+/;print "\t$F[1]"; ' \
> aa.dat
mysql -B -N -e 'select accession,organism from cmp_genomics.genomes where user=USER()' \
> organism.names
foreach accession (AE017042 AE016879 AL111168 AL645882 AP008232 AP009048 BA000021 CP000034)
cat ~/Ex3/source/$accession.codon-usage.aa.list \
| perl -ane 'BEGIN { foreach (`cat organism.names`) \
{chomp; @A = split (/\t+/,$_);$NAMES{$A[0]} = $A[1];} } \
next unless $F[1] =~ /^[A-Z]+/;$FREQ{$F[1]} = $F[3]; push @AA,$F[1]; \
END { print "\n"; print "$NAMES{'$accession'}"; foreach $aa (@AA){print "\t",$FREQ{$aa};}}' \
>> aa.dat
end
</pre>
<pre>
mysql -e "delete from localization where user=USER() and type='signalp'" cmp_genomics
foreach accession (AE017042 AE016879 AL111168 AL645882 AP008232 AP009048 BA000021 CP000034)
perl -ne 'if(/^CDS_(\d+)-(\d+)_DIR([+-])\s+0.\d+\s+\d+\s+[YN]\s+0.\d+\s+\d+\s+[YN]\s+[10].\d+\s+\d+\s+Y\s+[10].\d+\s+Y\s+[10].\s+\s+Y\s+CDS_\d+-\d+_DIR[+-]\s+Q\s+[10].\d+\s+[10]\s+Y\s+[10].\d+\s+Y/) \
{print "insert into localization (user, protid, type) \
select USER() as user, ft.id as protid, \"signalp\" as type \
from features as ft \
where ft.user=USER() and \
ft.accession = \"'$accession'\" and \
ft.start = $1 and ft.stop = $2 and \
ft.dir = \"$3\";\n";}' $accession.signalp \
| mysql -D cmp_genomics
end
</pre>
<pre>
perl -e 'foreach $i (1..11) { open IN , "col$i"; while (<IN>) { chomp;($a , $c) = split (/\t+/,$_); $DATA{$a}{$i} = $c;} close IN;} \
END {foreach $a ( reverse sort keys %DATA) {foreach $i (1..11) {$DATA{$a}{$i} = 0 if $DATA{$a}{$i} eq "" and $i > 1; \
print $DATA{$a}{$i}; print "\t" if $i < 11;}print "\n";}}' \
> final.dat

R
postscript ('final.ps')
source ( "/home/people/kiil/Exercises/files/heatmap.R" )
data <- as.matrix ( read.table ( 'final.dat' , row.names=1, header= TRUE,sep = "\t" ) )
pal <- maPalette(low = rgb(10,10,10,maxColorValue = 10), high = rgb(0,0,10,maxColorValue = 10), mid= rgb(5,5,10,maxColorValue = 10),, k = 1024)
par(oma=c(0, 0, 0, 0.5))
data <- scale(data)
new_heatmap ( data , cexCol=1,cexRow=1,colors = pal)
dev.off()
quit(save="no")
</pre>

@@ Line 148: / Line 148: @@
 *Change from one biological format to another:
   perl -MBio::SeqIO -e '$informat="genbank";$outformat="fasta";$count=0;for $infile (@ARGV){$in=Bio::SeqIO->newFh(-file =>$infile, -format =>$informat);$out=Bio::SeqIO->newFh(-format =>$outformat);while(<$in>){print $out $_;$count++;}}' foo.gbk >foo.fsa
 ==External links==
 *[http://sysbio.harvard.edu/csb/resources/computational/scriptome/UNIX/Tools/Change.html]

@@ Line 92: / Line 92: @@
 quit(save="no")
 </pre>
 ==Transpose matrix==

← Older revision		Revision as of 00:46, 27 April 2007
Line 114:		Line 114:
	if $unequal;warn "\nTransposed table: result has $. columns and $num_out_rows rows\n\n" }' $1		if $unequal;warn "\nTransposed table: result has $. columns and $num_out_rows rows\n\n" }' $1
	</pre>		</pre>
		+
		+	==Misc==
		+	*Change columns in each line of files:
		+	perl -e '$sep=",";while(<>){s/\Q$sep\E/\t/g;print $_;}' file.csv >file.tab
		+
		+	*Change tab-separated data to use a different delimiter:
		+	perl -e '$sep=",";while(<>){s/\t/$sep/g;print $_;}' file.tab >file.csv
		+
		+	*Remove spaces in a line:
		+	perl -e 'while(<>){s/ //g;print $_;}' foo
		+
		+	*Change all characters to upper case:
		+	perl -e 'while(<>){print uc($_);}' foo
		+
		+	*Change a FASTA file into tabular format:
		+	perl -e '$count=0; $len=0; while(<>){s/\r?\n//;s/\t/ /g;if(s/^>//){if($. !=1){print "\n"}s/ \|$/\t/;$count++;$_ .="\t";}else{s/ //g;$len +=length($_)}print $_;}print "\n";' foo.fsa >foo.tab
		+
		+	*Change a tabular file into FASTA format:
		+	perl -e '$len=0;while(<>){s/\r?\n//;@F=split /\t/,$_;print ">$F[0]";if(length($F[1])){print" $F[1]"}print "\n";$s=$F[2];$len+=length($s);$s=~s/.{60}(?=.)/$&\n/g;print "$s\n";}' foo.tab >foo.fsa
		+
		+	*Change from one biological format to another:
		+	perl -MBio::SeqIO -e '$informat="genbank";$outformat="fasta";$count=0;for $infile (@ARGV){$in=Bio::SeqIO->newFh(-file =>$infile, -format =>$informat);$out=Bio::SeqIO->newFh(-format =>$outformat);while(<$in>){print $out $_;$count++;}}' foo.gbk >foo.fsa
		+
		+	==External links==
		+	*[http://sysbio.harvard.edu/csb/resources/computational/scriptome/UNIX/Tools/Change.html]

@@ Line 91: / Line 91: @@
 dev.off()
 quit(save="no")
 </pre>