Filo

Moved to...

Tue, 12 Mar 2013 04:22:23 -0400

Nipype 0.3.4 release

Fri, 14 Jan 2011 05:50:46 -0500

I’m very happy to announce new release of nipype - the pipelining
framework that allows you to combine FSL, SPM, FreeSurfer, Nipy and
other neuroimaging packages. Among other changes (see the full list
below) this release brings support for fitting GLM models, estimating
contrasts and calculating brain masks from EPI sequences using nipy.

You can get the newest version of nipype from
https://github.com/nipy/nipype/zipball/0.3.4 and soon from NeuroDebian
repository ( http://http://neuro.debian.net/ ). Full documentation
with examples can be found at
http://nipy.sourceforge.net/nipype/documentation.html

Changes since 0.3.3:

* API: hash values for float use a string conversion up to the 10th
decimal place.
* API: Iterables in output path will always be generated as
_var1_val1_var2_val2 pairs

* ENH: Added support to nipy: GLM fit, contrast estimation and
calculating mask from EPI
* ENH: Added support for manipulating surface files in Freesurfer:
- projecting volume images onto the surface
- smoothing along the surface
- transforming a surface image from one subject to another
- using tksurfer to save pictures of the surface
* ENH: Added support for flash processing using FreeSurfer
* ENH: Added support for flirt matrix in BBRegister
* ENH: Added support for FSL convert_xfm
* ENH: hashes can be updated again without rerunning all nodes.
* ENH: Added multiple regression design for FSL
* ENH: Added SPM based Analyze to Nifti converter
* ENH: Added increased support for PyXNAT
* ENH: Added support for MCR-based binary version of SPM
* ENH: Added SPM node for calculating various threshold statistics
* ENH: Added distance and dissimilarity measurements

* BF: Diffusion toolkit gets installed
* BF: Changed FNIRT interface to accept flexible lists (rather than 4-tuples)
on all options specific to different subsampling levels

matplotlib can be a real memory hog

Thu, 11 Nov 2010 09:57:46 -0500

Always remember to do:

plt.clf()
plt.close()

when you finish with plotting/saving figures. See also:

http://stackoverflow.com/questions/3623600/python-matplotlib-memory-not-being-released-when-specifying-figure-size

and

http://stackoverflow.com/questions/2364945/matplotlib-runs-out-of-memory-when-plotting-in-a-loop-edited

Amazon can be hilarious

Sun, 22 Aug 2010 21:54:31 -0400

Customers who bought Life Nomadic also bought The Game and Grow Great Marijuana. So travelling is all about picking up girls and smoking weed?

NiPyPe: Neuroimaging in Python Pipelines and Interfaces

Tue, 03 Aug 2010 19:48:00 -0400

Dear members of the neuroimaging community,

We are pleased to announce release 0.3.2 of NiPyPe [nai-pipe]:
Neuroimaging in Python Pipelines and Interfaces
(http://nipy.org/nipype).

Nipype, an open-source, community-developed initiative under the
umbrella of NiPy, is a Python project that provides a uniform
interface to existing neuroimaging software and facilitates
interaction between these packages within a single workflow. Nipype
provides an environment that encourages interactive exploration of
algorithms from different packages (e.g., SPM, FSL, FreeSurfer), eases
the design of workflows within and between packages, and reduces the
learning curve necessary to use different packages. Nipype is creating
a collaborative platform for neuroimaging software development in a
high-level language and addressing limitations of existing pipeline
systems.

Nipype allows you to:

easily interact with tools from different software packages
combine processing steps from different software packages
quickly develop new workflows by reusing common steps from existing workflows
process data faster by running it in parallel on many cores/machines
make your research easily reproducible
share your processing workflows with the community

For example workflows see:
http://nipy.sourceforge.net/nipype/users/pipeline_tutorial.html

Nipype is a young community-based project and we welcome your inputs,
contributions and criticisms.
Release Notes

Please see: http://nipy.org/nipype/changes.html
Where to go for help

the user mailing list: nipy-user@googlegroups.com [subscribe]: http://groups.google.com/group/nipy-user

the developer mailing list: nipy-devel@neuroimaging.scipy.org [subscribe]: http://mail.scipy.org/mailman/listinfo/nipy-devel

on both of these lists please prefix the subject heading with [nipype]
to indicate a nipype relevant post.

—- The NiPy Community

"git reset is probably the most confusing command written by humans. I’ve been using Git for..."

Wed, 28 Jul 2010 11:13:51 -0400

“git reset is probably the most confusing command written by humans. I’ve been using Git for years, even wrote a book on it and I still get confused by what it is going to do at times.”

- gitref.org

Comments

Fri, 02 Jul 2010 13:12:18 -0400

Yay! Rejoice! I’ve added comments to my blog. Now you can express how boring/stupid my post are!

Managing academic papers in Ubuntu

Thu, 01 Jul 2010 18:51:00 -0400

Recently I was approached by one of my colleagues and asked to show what do I use to managed papers and citations. The guy was a fan of latex but (as I did some time ago) he missed reviewing and collaboration tools available in programs like OpenOffice or MS Word. I showed him zotero with its OpenOffice plugin plus some addition tweaks I use search through my papers. His enthusiasm encouraged me to write down a simple howto describing how to setup efficient environment for storing and searching through academic papers.

The core of this setup is Zotero - a plugin for Firefox that stores and manges papers. To install it you need to start Firefox go to http://www.zotero.org/ and click download.
Zotero is very simple to use - each time we go to a website recognized a paper (on pubmed or science direct for example) we’ll see a little icon in the far right side of the address bar. After clicking it the paper will automatically be imported into our collection. You can browse it by clicking on the zotero icon in the bottom right corner.
Zotero can export collection into many formats including bibtex, but it also provides plugins for OpenOffice and MS Word which allows you to directly insert citations into your documents. Those plugins need to be downloaded separately from http://www.zotero.org .
There are few options worth turning on. Go to preferences by clinking zotero icon and the cog icon. In General settings select “Automatically attach assosiated PDFs…”.
In addition there are two synchronisation options: metadata and pdfs. First one is done through zotero servers and is easily configured through Sync tab. For the second one we’ll need a free webdav account from http://www.box.net. After acquiring it go to Sync tab choose WebDAV instead of Zotero, “http”, “www.box.net/dav”, and email you used to register with box.net as the username. Now both the metadata and pdf will be synchronised through the cloud so, for example, you can access the same collection from home and work.
Now it’s time to tweak usability. For this we’ll need two components: a file indexer and a fast launcher. The first one is called “Tracker Search Tool” and he second one is called Gnome Do. Both of them can be found in the Ubuntu repository.
We need to configure the indexer. Go to System -> Preferences -> Search and Indexing. Choose Enable indexing. Go to Files. Deselect Index home folder. Add “~/.mozilla/firefox/ibif4bcr.default/zotero/storage/”. ibif4bcr.default will be something different for your case. If you have a folder with books and reference materials add it as well. Press ok. Tracker will start indexing your files.
Now the launcher. Accessories -> Gnome Do . Click on the triangle to open preferences. Enable Tracker Search community plugin.
After a while your tracker will finish indexing pdf imported using zotero (if you have any). Press Super (windows key) + Space. Gnome Do window will appear. Type the phrase you are looking for. Scroll with arrow to the bottom. Press Tab. Scroll to Search with Tracker. A list of related pdfs and other indexed files will appear. This is a very quick and easy way to find a paper.

Enjoy!

A Google Translate gem

Wed, 16 Jun 2010 09:47:45 -0400

Mają kurwa rozmach. == They have momentum whore.

At a bar near Times Sq

Wed, 16 Jun 2010 09:45:11 -0400

Me: Can I have a pint of ... please?
Bartender: Can I see your I.D.?
Me: Sure.
Bartender: Oh you're Polish! I know one sentence in Polish!
Me: Bring it on. Lets hear it!
Bartender: Kurwa mać.

I wasn't even surprised...

How to configure Sun Grid Engine for FSL under Ubuntu

Fri, 07 May 2010 16:31:07 -0400

This short howto describes how to set up a local SGE so all of the cores of your workstation could be used.

sudo apt-get install gridengine-master gridengine-exec gridengine-client gridengine-qmon
sudo qmon
Queue Control->Add
set New Host/Hostgroup to your host name and press the arrow
Set the Processors and Slots to an appropriate number (number of cores - 1 is the recommended value if you plan to use your workstation during the processing)
Back to main menu and to Host Configuration
Submit Host->Add your host name

To run a job using your local SGE the FSLPARELL=1 has to be set. For example:

FSLPARALLEL=1 bedpostx bedpostx -b 1000 -n 1 -j 1250 -s 25 -w 1.00

You can monitor your local grid using qmon (Job Control).

"An activity difference between two experimental conditions (as tested by a conventional t test)..."

Fri, 07 May 2010 12:08:39 -0400

“An activity difference between two experimental conditions (as tested by a conventional t test) necessarily implies that the experimental condition can be predicted from the activity (classification approach) with some accuracy greater than chance. Thus, the title of every fMRI study revealing an effect of experimental condition on brain activity (i.e. the vast majority of fMRI studies) could legitimately be changed to claim that the mental state associated with each condition can be “predicted” or decoded from measured brain activity.”

- N. Kriegeskorte et al. Neuroimage 38

"In San Antonio, Texas, a number of Mexican-American women participated in a 1971 study to determine..."

Thu, 29 Apr 2010 14:37:32 -0400

“In San Antonio, Texas, a number of Mexican-American women participated in a 1971 study to determine side effects of an oral contraceptive. The women came to a clinic seeking contraceptives. Unbeknownst to them, the study was designed so that half the women would receive oral contraceptives for the first half of the study, then switched to placebo. The women initially receiving placebo were placed on the oral contraceptive for the second half of the study. Ten of the 76 subjects became pregnant while using placebo.”

- Levine, Robert J. Ethics and Regulation of Clinical Research. 2nd ed. New Haven CT: Yale University Press, 1988

The buzzword of the day

Fri, 23 Apr 2010 18:20:23 -0400

being not factually accurate = bullshit

Awesome teabag!

Sat, 17 Apr 2010 09:06:12 -0400

Awesome teabag!

...

Wed, 14 Apr 2010 12:39:05 -0400

Ostatnie wydarzenia w moim rodzinnym kraju sprowokowały mnie do pewnych przemyśleń. W tragiczny sposób zmarło wielu polityków, których za ich życia nie darzyłem wielką sympatia. Byli to ludzie, którzy nie reprezentowali moich poglądów - nie głosowałem na nich. Tak jednak w demokracji jest, że biorąc udział w wyborach jesteśmy gotowi przyjąć ich wynik. Nieważne czy głosowaliśmy na zwycięzcę czy nie to on będzie reprezentował kraj i jego decyzjom (czy się nam to podoba czy nie) będziemy się podporządkowywać. Sobotnia tragedia nie miała, więc tylko wymiaru ludzkiego (choć to wystarczająco wstrząsające wydarzenie) - nagłej śmierci 96 ludzi, ale także wymiar symboliczny. Zginęła głowa państwa.

Tak się składa, że jak bardzo byśmy tego nie chcieli wartościujemy ludzkie życie. Śmierć jednej osoby wydaje się być większą stratą od śmierci innej. Oczywiście brzmi to okropnie, ale to naturalny ludzki odruch. Śmierć prezydenta była uznana za dużą stratę. Tym niemniej, jeszcze za jego kadencji pracował o w opozycji do partii rządzącej, zbierał dość kiepskie oceny w sondażach i ostatnio nie cieszył się większą popularnością (ba nawet do niedawna istniała strona odliczają dni do końca jego kadencji). Tragiczna śmierć wzniosła go jednak na piedestał, który nie do końca wydaje się być zasłużony (szczególnie biorąc pod uwagę domniemane okoliczności feralnego wypadku).

Nie chciałbym być źle zrozumiany. Współczuje rodzinom zmarłych. Ta katastrofa to ogromna tragedia. Z mojego prywatnego punktu widzenia, pod względem emocjonalnym, nie różni się od katastrofy liniowca z normalnymi (choć zapewne na swój sposób wspaniałymi) pasażerami. Nie znałem nikogo osobiście na pokładzie tego samolotu. Moje przywiązanie do symboli narodowych (do jakich możemy zaliczyć prezydenta), tez ze względu na emigracje jest dość słabe.

Co spowodowało ten nagły zwrot akcji? Wyrzuty sumienia wszystkich jego wrogów politycznych, którzy wróżyli mu źle (jestem jednak przekonany, że śmierci mu nikt nie życzył)?

Zachodnie media chwalą stabilność polskiej demokracji, na której działanie te wydarzenia zbyt mocno nie wpłynęły. Prawda jest jednak taka, że zginął mało popularny polityk pełniący głównie reprezentacyjną funkcje. Do tego jak już wspomniałem działał w opozycji do partii rządzącej a inny ludzie pełniący ważne funkcje będący na pokładzie byli w większości mianowani przez niego. Do tego nowe wybory i tak miały być za sześć miesięcy. Nie był to tak naprawdę duży cios pod względem funkcjonowania państwa.

Jedno jest jednak pewne - ta tragedia wpłynie na najbliższe wybory. Nadchodząca kampania na pewno będzie należała do nietypowych. Do tego strata dwóch naczelnych kandydatów i przejęcie obowiązków prezydenta przez trzeciego wcale nie wróży zrównoważonego wyboru. Nieważne jakie są moje prywatne opcje wyniki wyborów powinny odzwierciedlać opinie większości nieograniczone wypadkami losowymi.

Feels like home ;)

Fri, 09 Apr 2010 12:34:43 -0400

Feels like home ;)

New Englands weirdness

Wed, 07 Apr 2010 16:50:58 -0400

Me: Excuse me how can I get to Cambridge.
Passerby: You first need to get a train to Haymarket...

"It [the neurovascular coupling problem] is like a discussion between a plumber and an electrician."

Fri, 05 Mar 2010 11:54:34 -0500

““It [the neurovascular coupling problem] is like a discussion between a plumber and an electrician.””

- N. Logothetis

"Bonferroni’s inequalities are described in his 1935 article (a practical application) while a more..."

Sat, 09 Jan 2010 08:12:05 -0500

“Bonferroni’s inequalities are described in his 1935 article (a practical application) while a more theoretical treatment was released a year later. As usual in statistics, the name can be
considered inappropriate as the inequality used in statistical simultaneous inference relies only on Boole’s inequality. Indeed, this is an example of Stigler’s Law of Eponymy, which states that: “No scientific discovery is named after its original discoverer.” Stigler notes that he, himself, did not discover this phenomenon, and thus feels quite justified in applying his name to it.”

- Federico Turkheimer, John Aston, Vincent Cunningham