Version 3.0 ?

Help
Anonymous
2011-09-01
2013-04-25

  • Anonymous
    2011-09-01

    I would like to use ggsd for big amount of data in my laboratory.
    ggsd is installed and seems to work perfectly.
    I saw patches arriving last month and a new version seems to be
    in progress.
    Do you have any idea when this version will be available ?
    Will it be easy to move from 2.19 to 3.0 ?

     
  • Hi,

    Thanks for your interest in GGSD and I'm glad you are finding it useful thus far.

    You are correct that I'm in the process of making a new release of the software and it will be version 3.0.  The new version will have a new schema and will be using a new storage engine (InnoDB) in MySQL.  Due to the changes GGSD will require at least MySQL 5.1 and will no longer work with MySQL 5.0.  In addition I'm adding the capability to store analysis results and recoding to hopefully speed up the applicaiton.  My goal is to have the new version out in November.

    I'm not sure about the ease of moving between the versions.  But with the schema changes, code changes, and MySQL requirement changes you might need to re-upload all your data into the new system.  I'm happy to discuss this further once the time comes if you want.

    Best regards,

    Aaron

     
  • Hi Aaron,

    Are you still working on version 3.0 ?
    Any idea when it could be ready ?

    Best regards

    Anne Degavre

     
  • Hi Anne,

    I am still working on v3.0.  I know I said I wanted it out in November, but October was a bit of a lost month because of traveling to ICHG and travelling to visit collaborators after the conference.  I have finalized the new schema and need to finish coding the new functionality and updating the older scripts.  With upcoming travel and other project considerations, I would say a realistic date for the new release would be around christmas/new year.  Hopefully that will still work with your timelines.  I'm the only person coding/developing this, which restricts the speed of the development.

    Best,

    Aaron

     
  • Hi Aaron,

    I am happy to hear that you are still working on it. A release in the beginning of 2012 will be perfect for me.
    I will test it shortly after.
    Are you interested in any bug reporting ?
    I will have to make some modification in your code, as to accept other characters in pedigree Id's or perhaps add tables or fields. Are you interested in any feedback ?
    I couldn't find many posts on the site. Is there other teams using ggsd ?

    Best regards
    Anne

     
  • Hi Anne,

    I'm glad that this time frame will work for you.

    Yes, I'm always interested in hearing about all the bugs and suggestions for improvement that users have.  They will only make the application better.  If you will start submitting bug reports and feature requests through the sections in the "Tracker" tab at the top of the page I'll start fixing and incorporating into the new release.

    I'm developing the new version with/for my current group, but I'm not sure how many people are using the system or have tried it.  You've been the most communicative user thus far.

    Thanks for the interest and help identifying bugs and improvements.

    Best,

    Aaron

     
  • Hi Anne,

    I don't know if you'll get notified, but I've corrected the bug you submitted and checked in an update.  If you check the bug report it will tell you where you can download just the updated script.

    -Aaron

     
  • Hi Aaron,

    It looks like you are still working on GGSD. Do you have any idea when you are going to release version 3.0 ?
    I have been working on my data and I am still very interested.

    Best regards
    Anne

     
  • Hi Anne,

    Yes, I'm still plugging away at version 3.  I had to take several unexpected trips in December and January that delayed things.  Additionally, designing a replication study and an analysis plan for a large consortium has taken priority recently.  All of these has slowed my progress.  But I'm making good advancement now and expect to make an initial release mid-April.  If you would like to discuss anything offline (migration, etc) please feel free to contact me at ern_day@users.sourceforge.net.

    Best,

    Aaron

     
  • Hi Anne,

    Wanted to update you.  I moved countries last week to start a new position, which has put me releasing the new version back a little.  I have someone at the Sanger Institute starting to test the new version.  Once we put version 3 through some of its paces (and I fix the identified bugs) over the next two-three weeks I'll be releasing the new version.  In limited testing inserts are a lot faster than version 2.  Hope it will still be useful for you.

    Best,

    Aaron

     
  • Anne Degavre
    Anne Degavre
    2012-05-16

    Hi Aaaron,

    Do you mean you are changing of job and country ?
    Will you still be working on ggsd after releasing version 3.0 ?
    I am still interested of course !
    Thank you for giving me news.

    Best regards,
    Anne

     
  • Hi Anne,

    Yes, in the past 3 weeks I've moved countries and started a new position.  Which you can imagine has slowed me down on development.  Once version 3 is released, I'll definitely be maintaining the code and time permitting continue development.  I'll keep you updated.

    Best,

    Aaron

     
  • Hi Aaron,

    I just encountered snps that are insertions or deletions and I was wondering how you would
    insert that  in GGSD ?
    Allele size and sequence are identical for a change of base to G or an insertion of one G. Just changing the allele Code
    seems a bit dangerous for distractful users.

    Any news of version 3 ?

    Best
    Anne

     
  • Hi Anne,

    Re v3…..  When I did a demo for a couple of groups of v3 they both identified a need to allow individuals to have multiple ids (aliases).  This requires a reworking of the schema.  I'm working with someone at Sanger to alter the schema to allow this feature.  Once the schema is stable then the code needs to be updated.  I'll be making a big push on this over the next month.  I'm so sorry for the repeated delays, but v3 will have a lot more functionality and be more robust than v2.  Since I've been so bad about meeting suggested timelines and hesitant to give one.  I will say that it will be released before ASHG in November.

    As for dealing with insertions and deletions, I think GGSD will handle these just fine.  My thinking and development of the schema/code is based on the way they are represented in VCF format.  For example (taken from http://www.1000genomes.org/wiki/Analysis/Variant%20Call%20Format/vcf-variant-call-format-version-41):

    Example 1:  Consider the situation where the reference genome has sequence GCG starting at position 4 on chr 20, and you've observed a 2bp deletion (G) and a 2bp insertion (GCGCG).  Then in GGSD this would be represented as a microsatellite with the following information

    Marker Table Info:
    marker_name                 => micro1
    chromosome                  => 20
    chromosome_position => 4
    number_of_alleles        => 3

    Allele Table Info:
    Allele 1 (ref allele):
    allele_code => 1
    allele_sequence => GCG
    allele_size            => 3

    Allele 2 (del):
    allele_code           => 2
    allele_sequence => G
    allele_size            => 1

    Allele 3 (ins):
    allele_code          => 3
    allele_sequence => GCGCG
    allele_size            => 5

    Example 2:  Suppose you only have a one base insertion of a G after the reference base G at position 4 on chromosome 20.  This would be represented in GGSD as an insertion like:

    Marker Table Info:
    marker_name                => ins1
    chromosome                  => 20
    chromosome_position => 4
    number_of_alleles        => 2

    Allele Table Info:
    Allele 1 (ref):
    allele_code          => 1
    allele_sequence => G
    allele_size            => 1

    Allele 2 (ins):
    allele_code          => 2
    allele_sequence => GG
    allele_size            => 2

    Example 3: A one base deletion of the C at base position 5 of chromosome 20 (with G at position 4).  This would be represented in GGSD as a deletion like:

    Marker Table Info:
    marker_name                => del1
    chromosome                  => 20
    chromosome_position => 4
    number_of_alleles        => 2

    Allele Table Info:
    Allele 1 (ref):
    allele_code          => 1
    allele_sequence => GC
    allele_size            => 2

    Allele 2 (del):
    allele_code          => 2
    allele_sequence => G
    allele_size            => 1

    And I think that covers it.  Now it also seems you are asking about how to handle a substitution ("Allele size and sequence are identical for a change of base to G").  Isn't a substitution just a SNP (2 alleles: 1) original base 2) substituted base)? Or am I misunderstanding your question?

    I hope this helps and answers your questions.  Again, sorry for the slow down in the release of v3 but v3 will be expanded and more robust than v2. 

    Best,

    Aaron

     
  • Hi Aaron,

    I read your last reply and I am still not completely convinced but it is not easy to discuss the subject by mail.
    I find it more complicate to have an exhaustive vue of snps at a precise position.
    Concerning ggsd 3.0, I suppose you had too much ohter work to finish what you wanted to do and I was wondering if it could be possible to get a beta version. I could test it and give you a return on bugs and functionalities on a confidential basis of course.
    I hope this possibility will suit you and that we will quickly be able to discuss the details of it.

    Best Regards
    Anne Degavre
    INSERM
    France

     
  • Hi Anne,

    I'm happy with this arrangement.  I've sent you a personal email so we can take this offline and discuss.  Thank you for your understanding and interest in GGSD, it is helping keep me honest and moving forward with the development.  If I didn't find the correct email for you please email me at ern_day@users.sourceforge.net and we can start to discuss details.

    -Aaron