limiting gpSampler's maxStep

76 views
Skip to first unread message

Delaram

unread,
Apr 28, 2017, 6:49:19 AM4/28/17
to COBRA Toolbox
Hi,

I'm trying to use gpSampler function for a couple of small toy models. when I fix the upper and lower bounds of reactions on a certain flux value, I get  "warning small step", which leads to unreliable results.
In order to fix this problem,I tried changing  gpSampler optional input "maxSteps" from it's default value(1e10) to less values(down to 1e1).
I was wondering how this would effect the uniformity of sample points?

Thanks

Ronan M.T. Fleming

unread,
Apr 28, 2017, 7:00:47 AM4/28/17
to COBRA Toolbox

--

---
You received this message because you are subscribed to the Google Groups "COBRA Toolbox" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbox+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.



--
--
--
Mr. Ronan MT Fleming B.V.M.S. Dip. Math. Ph.D.
----------------------------------------------------------------------------
Senior research associate (EN) == Chercheur (FR),
Principal investigator,
Systems Biochemistry Group,
wwwen.uni.lu/lcsb/research/systems_biochemistry
Luxembourg Centre for Systems Biomedicine,
University of Luxembourg,
Campus Belval,
6, avenue du Swing,
L-4367 Belvaux.
&
National Centre of Excellence in Research on Parkinson’s disease
www.parkinson.lu
&
Adjunct Assistant Professor,
Division of Analytical Biosciences,
Leiden Academic Centre for Drug Research,
Faculty of Science,
University of Leiden.
http://analyticalbiosciences.leidenuniv.nl
----------------------------------------------------------------------------
Mobile:  +352 621 175 112
Office: +352 466 644 5528
Skype: ronan.fleming
----------------------------------------------------------------------------
(This message is confidential and may contain privileged information. It is intended for the named recipient only. If you receive it in error please notify me and permanently delete the original message and any copies.)

Delaram

unread,
Apr 28, 2017, 8:42:24 AM4/28/17
to COBRA Toolbox, ronan.mt...@uni.lu
Dear Ronan,
Thank you for your response.
I downloaded the document and tried sampling with chrrSampler but it gives the following error:

Error using nullSpaceOperator (line 96)
Chrr sampler not compatible with windows OS.

Error in getNullSpace (line 32)
nullS = nullSpaceOperator(S,gmscale,printLevel);        % forms a structure nullS.

Error in preprocess (line 32)
N = getNullSpace(P.A_eq,0);

Error in chrrSampler (line 127)
    [roundedPolytope] = preprocess(P, toRound);

Laurent Heirendt

unread,
Apr 28, 2017, 1:09:41 PM4/28/17
to COBRA Toolbox, ronan.mt...@uni.lu
Hi,

As the error states, the chrrSampler is not yet compatible with Windows OS. It can only be used on UNIX systems (macOS or linux) for now.

I hope this helps.

Best,

Laurent

Ronan M.T. Fleming

unread,
Apr 28, 2017, 1:16:09 PM4/28/17
to COBRA Toolbox
Replace

preprocess (line 32)
N = getNullSpace(P.A_eq,0);

with

archstr = computer('arch');
archstr = lower(archstr);
switch archstr
case {'win32', 'win64'}
N=null(P.A_eq);
case {'glnxa86'}
N = getNullSpace(P.A_eq,0);
end

It will be slower, but should still work.

--

Delaram

unread,
Apr 30, 2017, 4:44:32 AM4/30/17
to COBRA Toolbox, ronan.mt...@uni.lu
Unfortunately, It didn't work. This time I got the following errors: 

Error using svd
Use svds for sparse singular values and vectors.

Error in null (line 66)
[~,S,V] = svd(A,0);

Error in preprocess (line 36)
N=null(P.A_eq);

Ronan M.T. Fleming

unread,
May 2, 2017, 3:12:36 AM5/2/17
to COBRA Toolbox
archstr = computer('arch');
archstr = lower(archstr);
switch archstr
case {'win32', 'win64'}
N=null(full(P.A_eq));
case {'glnxa86'}
N = getNullSpace(P.A_eq,0);
end

Delaram

unread,
May 2, 2017, 4:41:02 AM5/2/17
to COBRA Toolbox, ronan.mt...@uni.lu

It worked this time !!!!!!  THANK YOU!



On Tuesday, May 2, 2017 at 11:42:36 AM UTC+4:30, Ronan M.T. Fleming wrote:
archstr = computer('arch');
archstr = lower(archstr);
switch archstr
case {'win32', 'win64'}
N=null(full(P.A_eq));
case {'glnxa86'}
N = getNullSpace(P.A_eq,0);
end
On 30 April 2017 at 10:44, Delaram <d.pouy...@gmail.com> wrote:
Unfortunately, It didn't work. This time I got the following errors: 

Error using svd
Use svds for sparse singular values and vectors.

Error in null (line 66)
[~,S,V] = svd(A,0);

Error in preprocess (line 36)
N=null(P.A_eq);

Error in chrrSampler (line 127)
[roundedPolytope] = preprocess(P, toRound);

--

---
You received this message because you are subscribed to the Google Groups "COBRA Toolbox" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cobra-toolbo...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages