August 2008 - Mg-rast - mailman3.cels.anl.gov

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Folker Meyer 29 Aug '08

29 Aug '08

sounds cool! FOlker On Aug 29, 2008, at 6:25 PM, Rob Edwards wrote: > The simplest way is to probably to set up an email verification > system. Automatically email the owner and say that we would like to > release their MG. If they agree click a link to authorize. No action > should be not authorized. Then we just need to record where the link > was clicked from and when it was clicked. That way it's simple and > we don't have to make any decisions other than to send them an email. > > Rob > > On Aug 29, 2008, at 9:05 AM, Folker Meyer <folker(a)anl.gov> wrote: > >> >> On Aug 29, 2008, at 6:03 PM, Robert Olson wrote: >> >>> I'll let you take that up with her :-). >> >> Please wait till I am back I want to videotape the occasion... >> >> ;-) >> >> >> >>> >>> >>> On Aug 29, 2008, at 11:01 AM, Rob Edwards wrote: >>> >>>> Perfect! VV should decide on the mg's too :) >>>> >>>> Rob >>>> >>>> On Aug 29, 2008, at 8:58 AM, Robert Olson wrote: >>>> >>>>> they all go thru veronika before going into the seed >>>>> >>>>> On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: >>>>> >>>>>> What is the example used in rast about release of genomes? >>>>>> Can't we follow that? >>>>>> >>>>>> Rob >>>>>> >>>>>> >>>>>> >>>>>> On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: >>>>>> >>>>>>> Hi all, >>>>>>> >>>>>>> I am getting really, really concerned about the possibility of >>>>>>> us "releasing" somebody's private data as public by accident. >>>>>>> >>>>>>> I believe we need some bookkeeping here that we can return to >>>>>>> if there are ever any questions. I suggest to log: >>>>>>> >>>>>>> - who made a metagenome public (basically one of us or the >>>>>>> submitting user) >>>>>>> - a clear text field that we can paste and email or URL into >>>>>>> that explains why a dataset is public >>>>>>> >>>>>>> It might be a good idea to always display this information, >>>>>>> but at least we should have that per data set. >>>>>>> >>>>>>> What are you thoughts on this? >>>>>>> >>>>>>> Folker >>>>>> >>>>> >>>> >>> >>

1 0

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Rob Edwards 29 Aug '08

29 Aug '08

The simplest way is to probably to set up an email verification system. Automatically email the owner and say that we would like to release their MG. If they agree click a link to authorize. No action should be not authorized. Then we just need to record where the link was clicked from and when it was clicked. That way it's simple and we don't have to make any decisions other than to send them an email. Rob On Aug 29, 2008, at 9:05 AM, Folker Meyer <folker(a)anl.gov> wrote: > > On Aug 29, 2008, at 6:03 PM, Robert Olson wrote: > >> I'll let you take that up with her :-). > > Please wait till I am back I want to videotape the occasion... > > ;-) > > > >> >> >> On Aug 29, 2008, at 11:01 AM, Rob Edwards wrote: >> >>> Perfect! VV should decide on the mg's too :) >>> >>> Rob >>> >>> On Aug 29, 2008, at 8:58 AM, Robert Olson wrote: >>> >>>> they all go thru veronika before going into the seed >>>> >>>> On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: >>>> >>>>> What is the example used in rast about release of genomes? Can't >>>>> we follow that? >>>>> >>>>> Rob >>>>> >>>>> >>>>> >>>>> On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: >>>>> >>>>>> Hi all, >>>>>> >>>>>> I am getting really, really concerned about the possibility of >>>>>> us "releasing" somebody's private data as public by accident. >>>>>> >>>>>> I believe we need some bookkeeping here that we can return to >>>>>> if there are ever any questions. I suggest to log: >>>>>> >>>>>> - who made a metagenome public (basically one of us or the >>>>>> submitting user) >>>>>> - a clear text field that we can paste and email or URL into >>>>>> that explains why a dataset is public >>>>>> >>>>>> It might be a good idea to always display this information, but >>>>>> at least we should have that per data set. >>>>>> >>>>>> What are you thoughts on this? >>>>>> >>>>>> Folker >>>>> >>>> >>> >> >

1 0

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Folker Meyer 29 Aug '08

29 Aug '08

On Aug 29, 2008, at 6:03 PM, Robert Olson wrote: > I'll let you take that up with her :-). Please wait till I am back I want to videotape the occasion... ;-) > > > On Aug 29, 2008, at 11:01 AM, Rob Edwards wrote: > >> Perfect! VV should decide on the mg's too :) >> >> Rob >> >> On Aug 29, 2008, at 8:58 AM, Robert Olson wrote: >> >>> they all go thru veronika before going into the seed >>> >>> On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: >>> >>>> What is the example used in rast about release of genomes? Can't >>>> we follow that? >>>> >>>> Rob >>>> >>>> >>>> >>>> On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: >>>> >>>>> Hi all, >>>>> >>>>> I am getting really, really concerned about the possibility of >>>>> us "releasing" somebody's private data as public by accident. >>>>> >>>>> I believe we need some bookkeeping here that we can return to if >>>>> there are ever any questions. I suggest to log: >>>>> >>>>> - who made a metagenome public (basically one of us or the >>>>> submitting user) >>>>> - a clear text field that we can paste and email or URL into >>>>> that explains why a dataset is public >>>>> >>>>> It might be a good idea to always display this information, but >>>>> at least we should have that per data set. >>>>> >>>>> What are you thoughts on this? >>>>> >>>>> Folker >>>> >>> >> >

1 0

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Robert Olson 29 Aug '08

29 Aug '08

I'll let you take that up with her :-). On Aug 29, 2008, at 11:01 AM, Rob Edwards wrote: > Perfect! VV should decide on the mg's too :) > > Rob > > On Aug 29, 2008, at 8:58 AM, Robert Olson wrote: > >> they all go thru veronika before going into the seed >> >> On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: >> >>> What is the example used in rast about release of genomes? Can't >>> we follow that? >>> >>> Rob >>> >>> >>> >>> On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: >>> >>>> Hi all, >>>> >>>> I am getting really, really concerned about the possibility of >>>> us "releasing" somebody's private data as public by accident. >>>> >>>> I believe we need some bookkeeping here that we can return to if >>>> there are ever any questions. I suggest to log: >>>> >>>> - who made a metagenome public (basically one of us or the >>>> submitting user) >>>> - a clear text field that we can paste and email or URL into >>>> that explains why a dataset is public >>>> >>>> It might be a good idea to always display this information, but >>>> at least we should have that per data set. >>>> >>>> What are you thoughts on this? >>>> >>>> Folker >>> >> >

1 0

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Rob Edwards 29 Aug '08

29 Aug '08

Perfect! VV should decide on the mg's too :) Rob On Aug 29, 2008, at 8:58 AM, Robert Olson wrote: > they all go thru veronika before going into the seed > > On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: > >> What is the example used in rast about release of genomes? Can't we >> follow that? >> >> Rob >> >> >> >> On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: >> >>> Hi all, >>> >>> I am getting really, really concerned about the possibility of us >>> "releasing" somebody's private data as public by accident. >>> >>> I believe we need some bookkeeping here that we can return to if >>> there are ever any questions. I suggest to log: >>> >>> - who made a metagenome public (basically one of us or the >>> submitting user) >>> - a clear text field that we can paste and email or URL into that >>> explains why a dataset is public >>> >>> It might be a good idea to always display this information, but at >>> least we should have that per data set. >>> >>> What are you thoughts on this? >>> >>> Folker >> >

1 0

Re: [mg-rast] Re: public metagenomes in MG-RAST v2
by Robert Olson 29 Aug '08

29 Aug '08

they all go thru veronika before going into the seed On Aug 29, 2008, at 10:19 AM, Rob Edwards wrote: > What is the example used in rast about release of genomes? Can't we > follow that? > > Rob > > > > On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: > >> Hi all, >> >> I am getting really, really concerned about the possibility of us >> "releasing" somebody's private data as public by accident. >> >> I believe we need some bookkeeping here that we can return to if >> there are ever any questions. I suggest to log: >> >> - who made a metagenome public (basically one of us or the >> submitting user) >> - a clear text field that we can paste and email or URL into that >> explains why a dataset is public >> >> It might be a good idea to always display this information, but at >> least we should have that per data set. >> >> What are you thoughts on this? >> >> Folker >

1 0

Re: public metagenomes in MG-RAST v2
by Rob Edwards 29 Aug '08

29 Aug '08

What is the example used in rast about release of genomes? Can't we follow that? Rob On Aug 29, 2008, at 6:01 AM, Folker Meyer wrote: > Hi all, > > I am getting really, really concerned about the possibility of us > "releasing" somebody's private data as public by accident. > > I believe we need some bookkeeping here that we can return to if > there are ever any questions. I suggest to log: > > - who made a metagenome public (basically one of us or the > submitting user) > - a clear text field that we can paste and email or URL into that > explains why a dataset is public > > It might be a good idea to always display this information, but at > least we should have that per data set. > > What are you thoughts on this? > > Folker

1 0

public metagenomes in MG-RAST v2
by Folker Meyer 29 Aug '08

29 Aug '08

Hi all, I am getting really, really concerned about the possibility of us "releasing" somebody's private data as public by accident. I believe we need some bookkeeping here that we can return to if there are ever any questions. I suggest to log: - who made a metagenome public (basically one of us or the submitting user) - a clear text field that we can paste and email or URL into that explains why a dataset is public It might be a good idea to always display this information, but at least we should have that per data set. What are you thoughts on this? Folker

1 0

Fwd: Your manuscript is acceptable for publication in principle.
by Folker Meyer 28 Aug '08

28 Aug '08

Here is the letter of acceptance. It came after I found out via the web page... Please update the web pages to: The metagenomics RAST server - a public resource for the automatic phylogenetic and functional analysis of metagenomes Folker Meyer, Daniel Paarmann, Mark D'Souza, Robert Olson, Elizabeth M. Glass, Mike Kubal, Tobias Paczian, Alexis Rodriguez, Rick Stevens, Andreas Wilke, Jared Wilkening and Robert A Edwards BMC Bioinformatics in print Folker Begin forwarded message: > From: BioMed Central Editorial <editorial(a)biomedcentral.com> > Date: August 28, 2008 6:15:18 PM GMT+02:00 > To: Dr Folker Meyer <folker(a)mcs.anl.gov> > Subject: Your manuscript is acceptable for publication in principle. > > Authors: Folker Meyer, Daniel Paarmann, Mark D'Souza, Robert Olson, > Elizabeth M. Glass, Mike Kubal, Tobias Paczian, Alexis Rodriguez, > Rick Stevens, Andreas Wilke, Jared Wilkening and Robert A Edwards > Title : The metagenomics RAST server - a public resource for the > automatic phylogenetic and functional analysis of metagenomes > Journal: BMC Bioinformatics > MS : 1951587053183573 > > Dear Dr Meyer, > > Peer review of your manuscript (above) is now complete, and we are > delighted, in principle, to accept the manuscript for publication in > BMC Bioinformatics. > > However before acceptance, our editorial production team needs to > check the format of your manuscript, to ensure that it conforms to > the standards of the journal. They will get in touch with you > shortly to request any necessary changes or to confirm that none are > needed. > > Many apologies for the delay in getting back to you with an > editorial decision. Shortly after we received your revised > manuscript, we contacted one of the reviewers for a re-review, whose > remaining concern was the testing of the computational power of > your server by uploading a large dataset. However, because of > confidentiality concerns, the reviewer asked us to upload their > dataset on their behalf. Please accept my apologies, as, after > failing to upload the dataset on your server, due to increased > workload during this month, I have not followed up the reviewer's > concern on time. Since then, the reviewer contacted us, informing us > that it was not their intention to delay the acceptance of the > manuscript and asking us to forward their message to you (I have > pasted the letter at the end of this message). > > Please accept again my apologies for the delay in getting back to > you with a decision. I have since forwarded the reviewer's requests > to the relevant department for consideration to place measures to > ensure that manuscripts do not suffer delays at these points. > > If you have any problems or questions regarding your manuscript, > please do get in touch. > > With best wishes, > > Anastasios > > Anastasios Koutsos PhD > Assistant Editor > BMC-series journals > > Tel: +44 (0)20 7631 9921 > Facsimile: +44 (0)20 7631 9923 > e-mail: editorial(a)biomedcentral.com > Web: http://www.biomedcentral.com/ > > Comments of the reviewer: > ----------------------------------- > Dear Anastasios, > > normally it would not be my job to enquire about this, but I would > like > to keep my reputation as a reviewer with BMC and we need to be fair to > the authors of the ms timewise. > > It has been 1 month since I asked whether you could inquire with the > authors about this multifasta upload problem and the only thing I > heard > was an automatic reminder that I should return my review in 10 days > (which I cant if I dont get my query answered). I dont know whats > going > on, but this is not fair anymore to the authors. Also I dont want to > go > into the books as a reviewer that always sends in reviews late. > > I wanted to do a very simple thing: upload a dataset to the server I > needed to test/review. However, this is apparently very difficult to > do > if I want to keep anonymity as a reviewer. I think BMC Bioinformatics > should have a system/method in place to do this (it is an bioinfo > journal, so this should not be the first time this happens) and I > shouldnt be wasting my time on trying to figure this out. > > All other changes to the ms. were fine, this upload was the only thing > needed to finish the review. I never intended to give the authors a > hard > time or 'stall' the ms. I just wanted to do a proper review of the > server. If its such an issue, then accept the ms as it stands now > and we > just need to trust the authors on their word. But this goes against > the > whole principle of peer review. > > Please forward this mail to the authors of the ms. so at least they > know > its not me that is delaying their ms. > > >

1 0

Re: [mg-rast] analysis question
by Jared Wilkening 28 Aug '08

28 Aug '08

Well we are currently not using any the tax_sim_best_by_iden_#### tables for the front end. That probably not a good thing but we have never really thought about when those tables should be used instead of the best_by_psc (evalue). If we are running in to space issues we should either decide to use them and look into other reductions or get rid of them for the moment. Jared On Thu, Aug 28, 2008 at 11:27 AM, Robert Olson <olson(a)mcs.anl.gov> wrote: > :-) > > I was looking at the postgres db on bio-big and it's currently consuming > 523G; 693G remaining on the local disk. at some point we may need to look > into policies for paging the live sims data into and out of the database. > > Do we know any better what parameter sets would be reasonable for > precomputing analyses for the initial browsing? > > --bob > > > On Aug 28, 2008, at 11:10 AM, Folker Meyer wrote: > > I am about to write stuff that uses it. please do not delete them :) >> >> Folker >> On Aug 28, 2008, at 6:06 PM, Robert Olson wrote: >> >> do any of the current analyses in the mg-rast2 system use the full hits >>> data, or have all moved to the best-hit tables? >>> >>> >>> >> >

1 0