1 Aug 2003 07:18
1 Aug 2003 01:08
Re: $_ assignment question
Paul Edlefsen <pedlefsen <at> systemsbiology.org>
2003-07-31 23:08:26 GMT
2003-07-31 23:08:26 GMT
My understanding has been that Bio::SeqIO->new(..) will surely use the correct method, while new Bio::SeqIO might use the new(..) method from the class that the code appears in, which would be incorrect. Have I misunderstood the issue? :Paul Josh Lauricha wrote: >On Thu 14:32, Jason Stajich wrote: > > >>my $obj = new Bio::SeqIO(..); >>instead of >>my $obj = Bio::SeqIO->new(..); >> >>On Thu, 31 Jul 2003, Hilmar Lapp wrote: >> >> >> > >And why is that bad? Just because it might confuse someone? I personally >prefer new Bio::SeqIO(..); mainly because I don't like the SomePackage-> >accessors, if something is a function it should be exported otherwise it >should be called as a member function on some object. Using ->new makes >it look like the package is an object. > >On a totaly off topic note, any reason why the list dislikes signed >e-mails? >(Continue reading)
1 Aug 2003 01:14
Re: distant thoughts from a dinosaur...
Paul Edlefsen <pedlefsen <at> systemsbiology.org>
2003-07-31 23:14:06 GMT
2003-07-31 23:14:06 GMT
I of course would love to attend and can't, but this is my only objection. Well, I *could* come if I could stow away somehow, but I'll be in Boston at the time... Otherwise, I think it's a fantastic idea. I attended some of those MAGE hackathons and they were not only a great success but great fun as well. :Paul Lincoln Stein wrote: >An excellent idea. I don't think that we can extend the GMOD meeting by a day >because people have already mostly made their reservations, but we have time >during the breakout groups to meet and plan a bioperl 2 designathon. > >Lincoln > >On Wednesday 23 July 2003 11:27 pm, Hilmar Lapp wrote: > > >>There is a GMOD mtg scheduled for Sep 15-16 in Berkeley. I thought that >>might be a good idea for that seed group of people to meet up and start >>thrashing out a bioperl2 model, since some (many?) of us may be going >>there anyway. >> >>And/or wouldn't this be a very good target for a sponsored hackathon, >>since 1) there'd be a clearly defined goal, which 2) could be achieved >>much better and much more efficient once the participating people are >>locked into a room for some time. >> >>Ahem, no, I'm not going to suggest a sponsor ...(Continue reading)
1 Aug 2003 01:36
Re: $_ assignment question
Josh Lauricha <laurichj <at> bioinfo.ucr.edu>
2003-07-31 23:36:03 GMT
2003-07-31 23:36:03 GMT
On Thu 16:08, Paul Edlefsen wrote:
> My understanding has been that Bio::SeqIO->new(..) will surely use the
> correct method, while new Bio::SeqIO might use the new(..) method from
> the class that the code appears in, which would be incorrect. Have I
> misunderstood the issue?
>
You may be entirely correct, however I've always been under the
impression they are interchangable. If they aren't, what would be an
example of the wrong function being called? The only thing I can figure
is where someone is calling a function from another package which has
been declared (or exported) in that scope... for instance...
package MyExample;
use Bio::PrimarySeq;
sub new {
my($class) = <at> _;
bless my $self = {}, $class;
$self->{seq} = new Bio::PrimarySeq;
# might this call be confused as MyExample->new(Bio::PrimarySeq);
# but I think Bio::PrimarySeq being a bare word prevents that...
# so maybe this:
$self->{seq} = new 'Bio::PrimarySeq';
# but thats wrong and shouldn't be protected against.
...
return $self;
}
(Continue reading)
1 Aug 2003 03:31
Re: Parsing a netblast file
Wes Barris <wes.barris <at> csiro.au>
2003-08-01 01:31:06 GMT
2003-08-01 01:31:06 GMT
Jason Stajich wrote: >>Through trial and error I have narrowed down the problem to the negative >>sign in the database details. Here is the section in question from a >>netblast result file: >> >>Database: All GenBank+EMBL+DDBJ+PDB sequences (but no EST, STS, GSS, >>or phase 0, 1 or 2 HTGS sequences) >> 1,819,241 sequences; -24,217,474 total letters > > > integer overflow. The number of letters in nt is > than the > largest signed number (2147483647) that an integer can represent. > > Looks like nt length is 8,782,847,770 - seems like it has been larger than > INT_MAX for a while, surprised they haven't updated their code. Do you > have the latest version of netblast on your machine? A bug report to NCBI > is probably a good idea if you are running the latest version Hi Jason, Thanks for responding. Yes, I am running the latest blastcl3 from the NCBI ftp site. I had already alerted NCBI to the problem (although I didn't understand the source of the problem until you pointed it out). Here is their response. It doesn't look like they are interested in fixing it: -------------------------- We have some back compatibility issue for the older client and would not be able to change this.(Continue reading)
1 Aug 2003 04:10
Re: Parsing a netblast file
Jason Stajich <jason <at> cgt.duhs.duke.edu>
2003-08-01 02:10:33 GMT
2003-08-01 02:10:33 GMT
Here is the patch it is pretty simple or you just need to grab the latest
version of blast.pm from CVS.
Index: Bio/SearchIO/blast.pm
===================================================================
RCS file: /home/repository/bioperl/bioperl-live/Bio/SearchIO/blast.pm,v
retrieving revision 1.42.2.9
diff -r1.42.2.9 blast.pm
273c273
< if( /\(([\d,]+)\s+letters.*\)/ ) {
---
> if( /\((\-?[\d,]+)\s+letters.*\)/ ) {
325c325
< if(
/^\s+([\d\,]+)\s+sequences\;\s+([\d,]+)\s+total\s+letters/){
---
> if(
/^\s+(\-?[\d\,]+)\s+sequences\;\s+(\-?[\d,]+)\s+total\s+letters/){
525c525
< } elsif ( /letters in database:\s+([\d,]+)/i) {
---
> } elsif ( /letters in database:\s+(\-?[\d,]+)/i) {
On Fri, 1 Aug 2003, Wes Barris wrote:
> Jason Stajich wrote:
>
> >>Through trial and error I have narrowed down the problem to the negative
> >>sign in the database details. Here is the section in question from a
> >>netblast result file:
(Continue reading)
1 Aug 2003 04:27
Re: Panel add_track -label option does not work
Wes Barris <wes.barris <at> csiro.au>
2003-08-01 02:27:24 GMT
2003-08-01 02:27:24 GMT
Palle Villesen (BiRC) wrote: > Hi, > > Just checked the script, and here it does change the label nicely. When > given 0, no label is printed above each segment. When given 1, the $name > is printed above each and when given a fixed value (like 2 or 'palle' - > either 2 or palle is printed above each segment). > > But I dop agree on quite bad documentation of the Bio::Graphics modules. > Took me a long time to figure out, and I'm still worried about simple > problems I can't solve. But check if you saved the perl between testing > of 0 and 1 - it should work... You weren't a Microsoft support person in a former life, were you?(Continue reading)Yes, I am saving the perl script between tests. My computer is plugged in and turned on. I have done some more testing. I have installed bioperl on several systems: Redhat 8.0 / Bioperl-1.2.2 / GD 2.07: "-label => 1" fails to produce names above the features. Redhat 7.2 / Bioperl-1.2.1 / GD 1.33: "-label => 1" fails to produce names above the features. Solaris 8 / Bioperl-1.2.2 / GD 1.33: "-label => 1" works properly (the name apears above each feature). There is definately something subtly wrong with either bioperl-1.2.x, Redhat installations, or some unknown dependency which silently only affects the drawing of names above the features.
1 Aug 2003 04:36
Re: Parsing a netblast file
Wes Barris <wes.barris <at> csiro.au>
2003-08-01 02:36:23 GMT
2003-08-01 02:36:23 GMT
Jason Stajich wrote:
> Here is the patch it is pretty simple or you just need to grab the latest
> version of blast.pm from CVS.
Thank you! It works like a charm.
>
> Index: Bio/SearchIO/blast.pm
> ===================================================================
> RCS file: /home/repository/bioperl/bioperl-live/Bio/SearchIO/blast.pm,v
> retrieving revision 1.42.2.9
> diff -r1.42.2.9 blast.pm
> 273c273
> < if( /\(([\d,]+)\s+letters.*\)/ ) {
> ---
>
>> if( /\((\-?[\d,]+)\s+letters.*\)/ ) {
>
> 325c325
> < if(
> /^\s+([\d\,]+)\s+sequences\;\s+([\d,]+)\s+total\s+letters/){
> ---
>
>> if(
>
> /^\s+(\-?[\d\,]+)\s+sequences\;\s+(\-?[\d,]+)\s+total\s+letters/){
> 525c525
> < } elsif ( /letters in database:\s+([\d,]+)/i) {
> ---
(Continue reading)
1 Aug 2003 05:16
Errors generated from example code in HOW-TO section
Wes Barris <wes.barris <at> csiro.au>
2003-08-01 03:16:37 GMT
2003-08-01 03:16:37 GMT
The HOW-TO section http://www.bioperl.org/HOWTOs/html/Graphics-HOWTO.html shows examples using this code fragment: while( my $hit = $result->next_hit ) { next unless $hit->significance < 1.0; my $feature = Bio::SeqFeature::Generic->new(-score => $hit->raw_score, -seq_id => $hit->name, If warnings are turned on, then the 2nd line above produces these errors: Argument "e-171" isn't numeric in numeric lt (<) at blasttoimg.pl line 49, <DATA> line 191. Argument "e-163" isn't numeric in numeric lt (<) at blasttoimg.pl line 49, <DATA> line 191. Argument "e-160" isn't numeric in numeric lt (<) at blasttoimg.pl line 49, <DATA> line 191. Argument "e-160" isn't numeric in numeric lt (<) at blasttoimg.pl line 49, <DATA> line 191. Argument "e-160" isn't numeric in numeric lt (<) at blasttoimg.pl line 49, <DATA> line 191. I would suggest that when parsing a blast file, if the evalue does not begin with a digit, a '1' (one) should be placed at the beginning of the reported evalue. -- -- Wes Barris E-Mail: Wes.Barris <at> csiro.au
1 Aug 2003 05:18
Report to Recipient(s)
<QOLOTUS05 <at> genelogic.com>
2003-08-01 03:18:15 GMT
2003-08-01 03:18:15 GMT
Incident Information:- Originator: bioperl-l-bounces <at> portal.open-bio.org Recipients: Bioperl Mailing List <bioperl-l <at> bioperl.org> Subject: [Bioperl-l] Errors generated from example code in HOW-TO section Generic Alert: Exception was 'Corrupt File' scanning attachment '/ embedded'
RSS Feed