[Neurobureau-hubs] next large group call

Sun Jul 10 09:40:30 PDT 2011

Dear Joshua et al,

What you describe makes me think of one of the idea initiated by Mike
Milham. Basically we would have a number of research centers which would
provide some free space for datasets and some free resources for
computation, with as many tools as possible. I think it's important to have
multiple data repository centers to have "mirrors" and cut down access time
in various part of the world. These mirrors can be linked to multiple
places, such as NITRC. The key idea is that you can use the data and the
resources, or even upload your own data to process it there, but then you
have to commit to make it public, at least at some defined point. That's a
viral model for open data (and tool) sharing. Of course, the sites would
have common protocols, and share their databases to a large extent so they
would be driving forces in that network. What's cool about that model is
that there is no need to install a virtual machine to share tools or data,
and no limit to the computational power you can access (if the centers are
plugged into high-performance computing facilities).

I am definitely working towards that at my level. As I mentioned to you, I
am a "pipeliner" and I have developed software specifically targeted at
pipeline development and deployment (PSOM <http://code.google.com/p/psom/>and
NIAK <http://www.nitrc.org/projects/niak/>, which are the octave/matlab
equivalent of NIPYPE <http://nipy.sourceforge.net/nipype/> and
NIPY<http://nipy.sourceforge.net/nipy/stable/index.html>).
There is also the project of Alan Evans, CBRAIN <http://cbrain.mcgill.ca/>,
which aims at providing a generic interface to download/upload databases,
and send processing on a grid of supercomputers. Currently CBRAIN interfaces
about 50000 cores, but this has not been negotiated for open access. I am
pretty sure it could though, if rules were defined. Instead of writing this
ridiculously long email, I should actually be writing a paper about a
generic system to deploy any pipeline coded using PSOM into CBRAIN. That's
how we (me, Sébastien Lavoie-Courchesne and François Chouinard-Decorte) have
preprocessed the ADHD200 dataset. Alan also applied for a 5 Pb data storage
(I am on that application) so, if that goes through, Montreal could be one
of the mirroring site. We could try to implement this computational
infrastructure at various sites of the neurobureau and make it happen. It's
all possible in a short time. It could maybe even get funded, but as Mike
pointed out that's not a necessity.

BTW, because this project fits a number of large efforts in the field, I
think it's important for us to be connected with the key senior players
somehow. That's partly what motivated me to propose this "senior advisory
board" (it could be beneficial beyond that particular point though). If we
can be some sort of proof of concept for a larger effort, good. Conversely,
if someone gets funded to do that, hopefully we'll be part of it.

Finally, re the polymath project, one of the Alex of the neurobureau wrote
something about that on the neurobureau
blog<http://theneurobureau.wordpress.com/2011/05/26/a-brave-new-world-of-open-science/>.
We don't actually need a big infrastucture for that. All we need is a "grand
challenge" as well as a dataset where this grand challenge can be resolved.
Actually, the grand challenge could be the target of the workshop Donald
suggested. I have an idea for a grand challenge, will do a separate email
about that.

Please let me know what you think,

Pierre Bellec, PhD
Chercheur adjoint
Département d'informatique et de recherche opérationnelle
Centre de recherche de l'institut de Gériatrie de Montréal
4565, Chemin Queen-Mary
Montréal (Québec)
H3W 1W5
Université de Montréal
http://simexp-lab.org/brainwiki/doku.php?id=pierrebellec
(001)(514) 340 3540 #3367

2011/7/10 joshua vogelstein <joshuav at jhu.edu>

> ah, thank you for bringing up this point! i agree that this is a very
> important discussion, and that there are gradations of openness. consider
> wikipedia: it is not the case that anybody can just write anything they
> want.  wikimedia established guidelines, users are expected to follow.   the
> same is true of arxiv.  i expect that we will want to establish similar
> guidelines establishing what we are open to, and what we are not.  and i
> imagine that those guidelines will be organic and grow with us as we learn
> more and get more capabilities.
>
> cheers, j
>
>
>
>
>
>
> --
> If it makes you feel better, please remember to consider humanity before
> doing stuff. Otherwise, please just have a nice day.
> openconnectomeproject.org
>
>
>
>
> On Sun, Jul 10, 2011 at 10:03 AM, MCLAREN, Donald <
> mclaren.donald at gmail.com> wrote:
>
>> Along the discussion of openess. I think one of the early decisions if
>> if there will be a new central data repository (a number already
>> exist) or if we will use an existing one or if we will use multiple
>> systems that are distributed. The other decision related to this is
>> whether it should be fully open -- anyone can add to it -- or OPEN
>> access -- where an 'elected' committe would review and vote on the
>> submission or processing request -- or OPEN access -- where the
>> storage owners would decide what is valuable enough to put on the
>> system.
>>
>> These two questions will go hand in hand.
>>
>> On Saturday, July 9, 2011, Michael Milham <milham01 at gmail.com> wrote:
>> > Folks,
>> > Sitting in back of car stuck in traffic...so figured I'd give my
>> standard 2 cents sideline view commentary. When thinking through what to do
>> with an organization to make it meaningful, it is important to identify what
>> are the unaddressed needs of the community and what are the services you can
>> provide. I think it is important that folks think through how you want to
>> position the neurobureau in the community that will make it both unique and
>> effective (I.e., capable of taking ideas and making them reality) for the
>> community.
>> > With respect to grants, do not underestimate the environment we are
>> heading into at the NIH...Tom Insel has been very open about the reality of
>> the times for NIMH and the challenges ahead for funding...same true for the
>> other institutes. Review will be harder than ever and more competitive.
>> Foundations and philanthropy will be very important. And so will tempering
>> expectations, and increasing distribution of work...when you see efforts
>> like the 1000 functional connectomes project, INDI and ADHD-200...those were
>> all done without dedicated funding...they are worth it, but do take a toll
>> on those executing them (Maarten will readily testify to this, as I am sure
>> Cameron will in his more recent efforts). One hopes their efforts can obtain
>> funding over time as they become established enough...but that is over time.
>> So, my point is efforts up front will likely be the product of folks working
>> overtime or gaining philanthropic support. Would set goals for neurobureau
>> to ensure feasibility.
>> > Hope that makes sense.
>> > With respect to openness...my limited view of the situation is that it
>> may feel closed in that folks see pins and branding all over the place...but
>> not much saying "email us here to become a member"...keeps people who do not
>> know the inner circle looking from the outside...don't think that is
>> intentional...and my view can be off.
>> > Traffic has lightened...so, I will sign off on that note.
>> > Sent from my iPhone
>> > On Jul 9, 2011, at 11:32 PM, Pierre Bellec <pierre.bellec at criugm.qc.ca>
>> wrote:
>> >
>> > @Donald
>> > Yes, yes, yes !!! I love the idea of a workshop where people would try
>> to create something, rather than listen to talks. We had actually started
>> discussing something along those lines. One other idea was to have half
>> participants "senior" (should know what they're doing) and the other half
>> wanting to learn on a technique/set of techniques. We would pair seniors and
>> juniors based on interests. So it would also be an educational workshop.
>> Another idea would be to have a number of artists joining to work on one or
>> several pieces around the theme of the workshop, in interaction with the
>> scientists. In the case of Nathalie for example, there could even be some
>> imaging experiments going on as part of the workshop. Finally, I believe
>> there should be pre-workshop meetings on the web to discuss the work before
>> the event. 3, 4 or even 5 days are too short to achieve something if it's
>> not carefuly planned.
>> >
>> > Cheers,
>> > Pierre Bellec, PhD
>> > Chercheur adjointDépartement d'informatique et de recherche
>> opérationnelle
>> > Centre de recherche de l'institut de Gériatrie de Montréal
>> > 4565, Chemin Queen-Mary
>> > Montréal (Québec)
>> > H3W 1W5Université de Montréal
>> > http://simexp-lab.org/brainwiki/doku.php?id=pierrebellec
>> > (001)(514) 340 3540 #3367
>> >
>> >
>> >
>> >
>> > 2011/7/9 MCLAREN, Donald <mclaren.donald at gmail.com>
>> >
>> > I'm stealing this idea from the Advanced Psychometrics Workshop (its
>> > partially funded by the NIH).
>> >
>> > Each year, we should pick a place -- somewhere unique -- and hold a
>> > small workshop. I'm thinking 30 people maximum where they would apply
>> > and we'd choose the people. At the workshop, there would be some talks
>> > and then we'd divide into 3-4 workgroups and analyze a dataset. From
>> > this one or more papers could be produced from each group.
>> >
>> > I should also point out, that all the papers could form a special
>> > issue (e.g. Brain and Behavior is having a special issue just on the
>> > papers from the Advanced Psychometrics Workshop this year).
>> >
>> > After a year or two, we could probably get some NIH funding. This
>> > would be a good starting point for building a research focused
>> > organization. I also think that once we get going, then it would be
>> > easiest to be driven by corporate money. I think there will be a lot
>> > of resistance from institutions in the US from letting faculty apply
>> > for grants through the NB. To much lost revenue from the indirect
>> > costs.
>> >
>> > Now, if its truly a research institution, with its own facility,
>> > that's another story. However, I think that is probably a number of
>> > years away.
>> >
>> > At least that is how it seems.
>> >
>> > Best Regards, Donald McLaren
>> > =================
>> > D.G. McLaren, Ph.D.
>> > Postdoctoral Research Fellow, GRECC, Bedford VA
>> > Research Fellow, Department of Neurology, Massachusetts General Hospital
>> and
>> > Harvard Medical School
>> > Office: (773) 406-2464
>> > =====================
>> > This e-mail contains CONFIDENTIAL INFORMATION which may contain
>> PROTECTED
>> > HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
>> > intended only for the use of the individual or entity named above. If
>> the
>> > reader of the e-mail is not the intended recipient or the employee or
>> agent
>> > responsible for delivering it to the intended recipient, you are hereby
>> > notified that you are in possession of confidential and privileged
>> > information. Any unauthorized use, disclosure, copying or the taking of
>> any
>> > action in reliance on the contents of this information is strictly
>> > prohibited and may be unlawful. If you have received this e-mail
>> > unintentionally, please immediately notify the sender via telephone at
>> (773)
>> > 406-2464 or email.
>> >
>> >
>> >
>> >
>> > On Sat, Jul 9, 2011 at 10:56 PM, Pierre Bellec
>> > <
>>
>> --
>> Best Regards, Donald McLaren
>> =================
>> D.G. McLaren, Ph.D.
>> Postdoctoral Research Fellow, GRECC, Bedford VA
>> Research Fellow, Department of Neurology, Massachusetts General Hospital
>> and
>>
>> Harvard Medical School
>> Office: (773) 406-2464
>> =====================
>> This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
>> HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
>> intended only for the use of the individual or entity named above. If the
>> reader of the e-mail is not the intended recipient or the employee or
>> agent
>> responsible for delivering it to the intended recipient, you are hereby
>> notified that you are in possession of confidential and privileged
>> information. Any unauthorized use, disclosure, copying or the taking of
>> any
>> action in reliance on the contents of this information is strictly
>> prohibited and may be unlawful. If you have received this e-mail
>> unintentionally, please immediately notify the sender via telephone at
>> (773)
>>
>> 406-2464 or email.
>> _______________________________________________
>> Neurobureau-hubs mailing list
>> Neurobureau-hubs at www.nitrc.org
>> http://www.nitrc.org/mailman/listinfo/neurobureau-hubs
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.nitrc.org/pipermail/neurobureau-hubs/attachments/20110710/fd1e7027/attachment-0001.html