Hi Filipe

Thanks for your suggestions. Currently those are not possible, so I shall
take a look. However, having separate files may not be possible at the
moment.

The flag for memory used in the PSU is usually set to:  -mx1500m

If you have this memory available to you, you could start with that.

Regards
Tim

On 4/2/08 15:39, "Filipe Garrett" <[EMAIL PROTECTED]> wrote:

> 
> Hi all,
> 
> This new features seem just what I was looking for. However I still have some
> doubts:
> 
> Is there a way to set the "Frame Line Features..." options at start-up
> (through 
> the options file)?
> 
> I've noticed that relating sequence with feature is only possible when the
> FASTA 
> sequences are inside the GFF file. This is not so handy when your handling big
> files, such as entire genomes. Is it possible to relate both things when you
> add 
> each entry separately?
> 
> I also have some memory problems when loading an entire genome + GFF
> annotations. How much memory do you think it would be necessary for artemis to
> run? 2Gb?
> 
> A minor change, when loading a multi-FASTA file, would also come in hand.
> Artemis, currently, interprets the whole '>' string as the sequence ID. It
> would 
> be nicer if it was just until the first SPACE (the standard behaviour). That
> way 
> it is possible to add some description.
> 
> thanks in adv,
> FG
> 
> 
> Tim Carver wrote:
>> Hi Filipe
>> 
>> It sounds like you are better off using the "Frame Line Features..." option.
>> This allows you to define what features you want displayed on the frame
>> lines. I think the other option is there for convenience and for when it is
>> suitable.
>> 
>> In the development version of Artemis you can have GFF3 with multiple
>> sequences and the coordinates for the features relating to each (as below).
>> E.g.
>> 
>> ## gff-version 3
>> ##
>> contig1 foo     CDS     3       32      .       -       .
>> ID=cds1;colour=8
>> contig1 foo3    CDS     3       32      .       +       .
>> ID=cds3;colour=8
>> contig2 foo2    CDS     17      19      .       +       .
>> ID=cds2;colour=2
>> ###
>> ##FASTA
>>> contig1
>> gatcccaactgctcctaccgtcattggatccagcggcgttccctcttccttctccaccgt
>> tccttctagcagcggcgtcaagagctctaccaagactgcctcgaccgcaactgccaccga
>>> contig2
>> gatcccaactgctcctaccgtcattggatccagcggcgttccctcttccttctccaccgt
>> tccttctagcagcggcgtcaagagctctaccaagactgcctcgaccgcaactgccaccga
>> 
>> 
>> Hope this helps.
>> Regards
>> Tim
>> 
>>  
>> On 1/2/08 16:17, "Filipe Garrett" <[EMAIL PROTECTED]> wrote:
>> 
>>> Hi all,
>>> 
>>> I've been trying to use Artemis with a recent Ensembl files from Anopheles
>>> (3.46).
>>> 
>>> However, some things came to my atention.
>>> When I activate the "All features On Frame Lines" every feature is depicted
>>> on
>>> its correspondent lane. But why are exons also places on the frame lines?
>>> The
>>> exons don't have any frame! Shouldn't the frameless features be maintained
>>> on
>>> the central lines (or on a non-frame line)?
>>> The same thing happens when adding a multi-FASTA file. A concatenated
>>> sequence
>>> is loaded and several features, labelled "fasta_record", are created. Again
>>> this 
>>> features are placed as having frame!
>>> Is this a bug or am I forgetting something?
>>> 
>>> Another thing that I've noticed is that there is no use for the "seqname"
>>> field 
>>> in the GFF files. When placing features only the position and strand are
>>> taken
>>> into account, with all of them being placed in its positions but counting
>>> from
>>> the beginning of the concatenated sequence.
>>> Is there a way to link the "seqname" field with the "fasta_record" feature
>>> so
>>> that each feature is placed on the correct sequence? The idea would be to
>>> load
>>> a 
>>> single GFF with the annotations for all the sequences present in a
>>> multi-FASTA.
>>> 
>>> SEQ1 EMBL atg 103 105 . + 0
>>> SEQ1 EMBL cds 103 171 . + 0
>>> SEQ1 EMBL stop 172 174 . + .
>>> 
>>> SEQ2 EMBL atg 173 175 . + .
>>> SEQ2 EMBL cds 173 258 . + .
>>> SEQ2 EMBL stop 259 261 . + 0
>>> 
>>> 
>>> thanks in adv,
>>> FG
>>> 
>>> _______________________________________________
>>> Artemis-users mailing list
>>> [email protected]
>>> http://lists.sanger.ac.uk/mailman/listinfo/artemis-users
>> 
>> 
>> 
>> 
> 
> _______________________________________________
> Artemis-users mailing list
> [email protected]
> http://lists.sanger.ac.uk/mailman/listinfo/artemis-users



_______________________________________________
Artemis-users mailing list
[email protected]
http://lists.sanger.ac.uk/mailman/listinfo/artemis-users

Reply via email to