Basis sets, water scaling, baseline correction and data formats

[Paul]

Hi Martin and Greg,

First let me say thanks for TARQUIN, I have been analysing some data
with it (steam and MEGA-PRESS) and have been impressed so far. In
doing so I have come up with a few question you guy's might be able to
help me with.

The first is to do with basis sets. While the metabolites included in
the basis set are fairly extensive, would it be possible to add
extras? I assume to do so one would need to make a met.csv file that
lived in the basis set directory that contained the information about
spins,J coupling and frequencies just like the current metabolites
do. Would I be correct in that assumption? i.e if I create a
met.csv file and stick it in that directory, TARQUIN will include that
metabolite in the basis set, or is it a bit more complicated then
that? (I would like to include Glutathione among other things if
possible).

The second is to do with water scaling. Is it possible to change the
WATERconc value and the WATERatt value by setting a parameter
somewhere? This may help with fitting of data from different TE's or
tissue types if we could set the att factor either individually, or
just set it att to 1 and conc to 0.555 and do the relaxation and
tissue correction afterwards. I can of course do this afterwards in
excel or some other spreadsheet package, but if I could set these
parameter straight away it would save me some hassle redoing a bunch
of spreadsheets to take this into account. It might also be useful
for different data sets.

Thirdly, how is TARQUIN dealing with the baseline? or to put it
another way what method is it using for the baseline correction? I'm
curious more then anything, but it would be useful to know.

Lastly I have a question about data formats. I believe that right now
Tarquin won't take multi-voxel, or multi time series data from most
systems (or am I wrong?). In particular I believe it won't work well
with Philips SDAT/SPAR files that have more then one spectra. My
question is what format should I change this data too so that after I
average different time points I can use Tarquin to fit the data?
Would I need to convert the data to an LC-Model like raw format, or is
there another format that might work? (e.g basic text format output by
jmrui?) This would be really useful if I could do this, for a number
of reasons from attempts at functional MRS to some multi echo
fitting. In a similar vein, if I had some MEGA-PRESS data where the
fid pairs produced by the Philips system represent different events,
and I was able to convert and separate out all those pairs from the
same condition, to fit them with Tarquin, do they need to be in the
SDAT/SPAR format, or would the MEGA-PRESS option work with a simpler
data format?

Okay I think that's enough questions for the time being. If all my
questions are answered out in a paper or two that you have already
published, then just send me those papers or the reference and Ill
look them up. :) Actually if you have them and could send me PDF's or
the references for TARQUIN, please send me those anyhow so I can make
sure to reference it in the future.

Cheers
Paul.

[Martin Wilson]

Hi Paul,

Thanks for the feedback and questions, always good to have an idea of
what people are using TARQUIN for and how we can improve it.

Basis set question:
The default behaviour of TARQUIN is to use an "internal" basis set.
To add extra metabolites the easiest thing to do is to create a
directory containing the csv descriptions (an example is provided with
the download) and specify the Basis Set CSV directory in the advanced
fit dialogue. This should override the "internal" basis set. The
command line the option is --basis_csv. Creating new metabolite csv
files is a bit complex (single peaks are easier) so it might be best
to send me the information about the j-couplings and chemical shifts
and I can make one up for you. I have a Glth csv file derived from
the Govindaraju paper I'll send direct to you.

Water scaling question:
At the moment you can't change these values directly in the gui.
However you can specify a parameter text file in the advanced fit
dialogue that contains the following lines:

w_conc 1.0
w_att 1.0

to change from the default values. The same options can also be used
in the command line mode ie --w_conc. I'll add these to the next
version of the documentation.

Baseline question:
In short, the algorithm doesn't consider the first few points in the
FID when fitting. This means that rapidly decaying signals (ie
baseline) don't influence the fit. Once the FID is fitted the signal
is back-extrapolated over the ignored section and the difference
between the acquired signal and fit is regarded as baseline. Finally
a smoothing function is used in the frequency domain to visualise the
baseline after fitting. You can more details from the original paper:

http://tarquin.sourceforge.net/documentation.php ref MRM 2011 65(1):1-12.

Data formats question:
TARQUIN can read multi time series data from SPAR/SDAT and do an
average over all time points using the "press-all" pulse sequence
option. To extract averaged fid pairs use the "press-even" or
"press-odd" options. If you want to combine your data in more complex
ways it may be better to preprocess in jMRUI for now. We don't read
the jMRUI data format but can add support if it's simple enough.

I think I know what you have in mind for the MEGA-PRESS time series.
If you want TARQUIN to do the combination for you, then send me an
example SPAR/SDAT data file directly, explain which scans you want
averaging, and I'll have a go at implementing what you want.

Martin

[Pom Sailasuta]

Hi,
I have GE data sets for GABA

editing as well. Will send you data format to you to think about. Format is a two echo type.

Thanks for considering.

Pom

[Paul Mullins]

Thanks for the reply Martin. I'll try to get some data to you on Monday. Cheers.

Sent from my iPhone

[Martin]

Hi Pom,

The next version should work with the two echo MEGA-PRESS sequence
(Ruth and Ralph have given me some advice here), but more example data
is always useful for testing so thanks.

Martin

> >http://tarquin.sourceforge.net/documentation.phpref MRM 2011 65(1):1-12.

[Lotfi Kesraoui]

Hi Martin, 

Thank you again for Tarquin, it's wonderful. 

I don't find the files for running Tarquin with a command line. I am using a Mac. I downloaded this version TARQUIN_OSX_4.3.10.dmg but inside there are only the TarquinGui and two directories : basis_sets and example_data. 

Thanks ! 

[Paul Mullins]

Hi Lofti,

If you add lines like this:

PATH=/Applications/Tarquin/tarquingui.app/Contents/MacOS::${PATH}

export PATH

to your .bash_profile file in Mac OS you should then be able to run the tarquingui and tarquin commands from the command line. (As long as you have the tarquin.app file in a folder called Tarquin within your Applications folder

You can also just use this one line:

export PATH=/Applications/Tarquin/tarquingui.app/Contents/MacOS:$PATH

to do the same thing. 

Basically you need to add the tarquingui.app/Contents/MacOS/ directory to your path.

On Wed, Feb 8, 2017 at 7:20 PM, Lotfi Kesraoui <lotfik...@gmail.com> wrote:

Hi Martin, 

Thank you again for Tarquin, it's wonderful. 

I don't find the files for running Tarquin with a command line. I am using a Mac. I downloaded this version TARQUIN_OSX_4.3.10.dmg but inside there are only the TarquinGui and two directories : basis_sets and example_data. 

Thanks ! 

-- 

You received this message because you are subscribed to the Google Groups "TARQUIN users' group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tarquin_users_group+unsub...@googlegroups.com.
To post to this group, send email to tarquin_users_group@googlegroups.com.
Visit this group at https://groups.google.com/group/tarquin_users_group.
For more options, visit https://groups.google.com/d/optout.