[Bioperl-l] Errors using Bio::Assembly

[Andy Nunberg]

Hi, sorry this is so long, but Robson and myself have been trying to
determine why i am getting some warnings
when using the Bio::Assembly module

Please extract the code below and run it on the ace file in
bioperl-live/t/data/consed_project/edit_dir/test_project.fasta.screen.ace.1
to see if you get the same warnings I got.

At 11:53 PM 2/18/2003 -0300, you wrote:
>
>	Andy...
>
>> I would like to focus on this behavoir first, if thats ok with you.
>> try running this script using bioperl-1.2 on the same ace file and see what
>> you get, ok?
>> Andy
>
>	All right, but... the mistery remains: your code worked on my
>system. No errors, no warnings, same input file, both on Debian GNU/Linux
>and Tru64. By the way, I just noticed that my perl version is 5.6.1...
>check yours with perl -V and try upgrading if it is not 5.6.1 (although I
>guess it wont change anything...)
>	Anyway, based on your message below, I believe your error shows up
>while creating the contig object, because, at this moment, features
>holding read coordinates are added to the Bio::SeqFeature::Collection
>which is part of contig objects.
>	Maybe we should take the discussion back to the mailing list, to
>get someone's help on perl itself?
>						Robson
>On Tue, 18 Feb 2003, Andy Nunberg wrote:
>
>> Ok..
>> I am on a linux system, RedHat 7.2 using perl 5.6
>> there is no ace file in bioperl-live/t/data
>> well there is but it is not a phrap ace file i think its  a dbace file.
>> there is an ace file in /t/data/consed_project/edit_dir called
>> test_project.fasta.screen.ace.1
>> 
>> i used that ace file with teh following script
>> #!/usr/bin/perl -w
>> use strict;
>> use lib '/loginhome/anunberg/bioperl-1.2';
>> use Bio::Assembly::IO;
>> 
>> my $file=$ARGV[0];
>> 
>> my $in=Bio::Assembly::IO->new(-file=>$file, -format=>'ace');
>> warn "getting assembly...\n";
>> my $assembly=$in->next_assembly;
>> 
>> this is the output i got
>> getting assembly...
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 108.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 108.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 108.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 108.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 128.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 128.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 128.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 128.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 129.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 129.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 129.
>> Argument "" isn't numeric in numeric comparison (<=>) at
>> /loginhome/anunberg/bioperl-1.2/Bio/SeqFeature/Collection.pm line 436,
>> <GEN0> line 129.
>> 
>> 
>> At 02:32 PM 2/18/2003 -0300, you wrote:
>> >
>> >	Hi Andy,
>> >
>> >On Tue, 18 Feb 2003, Andy Nunberg wrote:
>> >> Rob,
>> >> Did you get my last e-mail?
>> >
>> >	No... and that is a little bit odd... :-)
>> >
>> >> I used your ace file and still got the errors
>> >> from Bio::SeqFeatures::Collection,
>> >> so maybe its my version of perl? (5.6)
>> >> Andy
>> >
>> >	No, it can't be that. I'm using the same perl version. Well, maybe
>> >it's your operanting system. I see from your error messages that you must
>> >be using some flavor of Unix. I expect my code to work on any Unix
>> >plataform but other plataforms may fail, because I defined $/ in
>> >next_assembly to be '\n', which is not used by Windows, for example. Are
>> >you running your scripts on Mac X? I've no experience with this plataform
>> >and things may not work on it.
>> >	Hmmmm... we are taking too long to solve this... lets try another
>> >approch. I'm sending you a few example scripts that I've made while
>> >developing Bio::Assembly modules and the output I get on my system when I
>> >run then. My systm is an Alpha running Tru64 Unix. Run the modules on your
>> >system and compare the results with the files I'm sending you, using diff.
>> >If you find any clues to the problem, tell me. Otherwise you may send the
>> >output of diff to me. Remeber to change the path to bioperl inside the
>> >scripts. The commands I used were
>> >
>> >bio_contig.pl > ~/bio_contig.out
>> >bio_assembly.pl > ~/bio_assembly.out
>> >bio_phrap.pl -f acefile.ace.1 > ~/bio_phrap.out
>> >bio_finishing.pl -f acefile.ace.1 -ncbs -lcq 25 -c 52 53 > 
>~/bio_finishing.out
>> >
>> >the ace file is inside t/data from bioperl-live's cvs repository.
>> >At the very least, bio_contig.pl and bio_assembly should run without
>> >errors. If your sequences show up on diff's output, just tell me what
>> >happened.
>> >			Robson
>> >
>> >PS: by the way, you can unpack .zip files can't you? :)
>> >
>> 
>> *******************************************************************
>> Andy Nunberg, Ph.D
>> Computational Biologist
>> Orion Genomics, LLC 
>> (314) 615-6989
>> http://www.oriongenomics.com
>> 
>> 

*******************************************************************
Andy Nunberg, Ph.D
Computational Biologist
Orion Genomics, LLC 
(314) 615-6989
http://www.oriongenomics.com