The reconstruction is in a constant state of flux, and some alterations have been proposed to the current version …
- r_0779, ATP + UDP <==> ADP + UTP : we should delete this reaction (but this would stop growth)
- r_0009, an 1-acyl-sn-gylcerol-3-phosphate acyltransferase catalysed by YDL052C: consider adding genes YOR175C, YGL084C, YPL189W as described at SGD
- r_1040, a triacylglycerol lipase catalysed by YKR089C|YMR313C|YOR081C: consider the multifunctional role of these genes as described at SGD
- add lipoic acid synthesis, as described in doi:10.1074/jbc.R800068200 and doi:10.1074/jbc.M109.015594
- use non-integral stoichiometries from iMM904 and iIN800
- fix any un-/under- annotated species
- look at reactions involving YBR196C
Please let us know of any other changes you would like to see.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Has anything been done to fix the non-boundary metabolites/boundary metabolites order? Half the model gets left out when you load it into cobra. What exactly is a boundary metabolite anyway?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Which version of Yeast and which version of COBRA are you using? Boundary metabolites have been addressed in the development version of Yeast since ~v4.30.
A boundary metabolite is a metabolite which is added to the model from outside - any metabolite that gets consumed. Without boundary metabolites, there would be no steady-state flux through the reaction network, since they participate in source reactions of the form -> A (as opposed to transport reactions like A(cyt) -> A(mito), or reactions like A+B -> C).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I thought I posted this twice, but my posts haven't been appearing somehow. I was using yeastnet v. 4.38 and the latest version of Cobra. It seems that in yeastnet, there are many metabolites labeled with an underscore b, "_b" that are not in the Boundary compartment. For instance:
Also, for metabolites that should have one boundary instance, all versions of it are labeled with underscore b.
Line 38181: <species id="s_0763_b" name="H+ " speciesType="t_0408" compartment="c_01" boundaryCondition="true"/>
Line 38182: <species id="s_0764_b" name="H+ " speciesType="t_0408" compartment="c_02" boundaryCondition="true">
Line 38189: <species id="s_0765_b" name="H+ " speciesType="t_0408" compartment="c_03" boundaryCondition="true">
Line 38196: <species id="s_0766_b" name="H+ " speciesType="t_0408" compartment="c_05" boundaryCondition="true" constant="true">
Line 38203: <species id="s_0767_b" name="H+ " speciesType="t_0408" compartment="c_06" boundaryCondition="true">
Line 38210: <species id="s_0768_b" name="H+ " speciesType="t_0408" compartment="c_08" boundaryCondition="true"/>
Line 38211: <species id="s_0769_b" name="H+ " speciesType="t_0408" compartment="c_10" boundaryCondition="true">
Line 38218: <species id="s_0770_b" name="H+ " speciesType="t_0408" compartment="c_12" boundaryCondition="true">
Line 38225: <species id="s_0771_b" name="H+ " speciesType="t_0408" compartment="c_14" boundaryCondition="true">
Line 38232: <species id="s_0772_b" name="H+ " speciesType="t_0408" compartment="c_16" boundaryCondition="true">
Line 38239: <species id="s_2884_b" name="H+ " speciesType="t_0408" compartment="boundary" boundaryCondition="true" constant="true"/>
In my understanding, only the last instance should have boundaryCondition = "true" and the underscore b.
In line 47 of the cobra toolbox function convertSBMLtoCobra, the code looks for the underscore b's to leave those metabolites out of the model.
% Ignore exchange species
if (isempty(regexp(modelSBML.species(i).id,'_b$','ONCE')));
Again, the model runs in cobra, but it is missing many metabolites, if this is how it is supposed to happen please explain, because currently I can't make any sense of the way this is set up, and it seems to me like your SBML file is being assembled by code with a bug in it.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Ah. Yep, that's wrong, and is fixed in a newer version (maybe even 4.39 or 4.40). Silly disagreements between the SBML standard and the COBRA toolbox.
Kieran and I are in the middle of trying to finalize changes for Yeast 5 for release in the next month or two. I had asked him to hold on to the recent updates pending publication. email me at bheavner at gmail and I'll send you a more recent version, but with the warning that it's got shortcomings, too.
4.05 was the last "stable" release, so everything since that point is in active development - things are changing around a lot at the moment. We're hoping to stabilize at 5.0 in the next month or two.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The reconstruction is in a constant state of flux, and some alterations have been proposed to the current version …
- r_0779, ATP + UDP <==> ADP + UTP : we should delete this reaction (but this would stop growth)
- r_0009, an 1-acyl-sn-gylcerol-3-phosphate acyltransferase catalysed by YDL052C: consider adding genes YOR175C, YGL084C, YPL189W as described at SGD
- r_1040, a triacylglycerol lipase catalysed by YKR089C|YMR313C|YOR081C: consider the multifunctional role of these genes as described at SGD
- add lipoic acid synthesis, as described in doi:10.1074/jbc.R800068200 and doi:10.1074/jbc.M109.015594
- use non-integral stoichiometries from iMM904 and iIN800
- fix any un-/under- annotated species
- look at reactions involving YBR196C
Please let us know of any other changes you would like to see.
The COBRA toolbox requires that non-boundary metabolites are placed before boundary metabolites. Grudgingly implement this for portability.
Hey Kieran - not sure if this is the best place for minor typos, but in the following, there's an extra comma:
<reaction id="r_0540" name="glycine hydroxymethyltransferase," sboTerm="SBO:0000176">
the name ends with a ",".
Has anything been done to fix the non-boundary metabolites/boundary metabolites order? Half the model gets left out when you load it into cobra. What exactly is a boundary metabolite anyway?
Hi Matthew,
Which version of Yeast and which version of COBRA are you using? Boundary metabolites have been addressed in the development version of Yeast since ~v4.30.
A boundary metabolite is a metabolite which is added to the model from outside - any metabolite that gets consumed. Without boundary metabolites, there would be no steady-state flux through the reaction network, since they participate in source reactions of the form -> A (as opposed to transport reactions like A(cyt) -> A(mito), or reactions like A+B -> C).
Hey Ben,
I thought I posted this twice, but my posts haven't been appearing somehow. I was using yeastnet v. 4.38 and the latest version of Cobra. It seems that in yeastnet, there are many metabolites labeled with an underscore b, "_b" that are not in the Boundary compartment. For instance:
<species id="s_1612_b" name="TSC10 " speciesType="t_0900" compartment="c_03" boundaryCondition="true" constant="true"/>
Also, for metabolites that should have one boundary instance, all versions of it are labeled with underscore b.
Line 38181: <species id="s_0763_b" name="H+ " speciesType="t_0408" compartment="c_01" boundaryCondition="true"/>
Line 38182: <species id="s_0764_b" name="H+ " speciesType="t_0408" compartment="c_02" boundaryCondition="true">
Line 38189: <species id="s_0765_b" name="H+ " speciesType="t_0408" compartment="c_03" boundaryCondition="true">
Line 38196: <species id="s_0766_b" name="H+ " speciesType="t_0408" compartment="c_05" boundaryCondition="true" constant="true">
Line 38203: <species id="s_0767_b" name="H+ " speciesType="t_0408" compartment="c_06" boundaryCondition="true">
Line 38210: <species id="s_0768_b" name="H+ " speciesType="t_0408" compartment="c_08" boundaryCondition="true"/>
Line 38211: <species id="s_0769_b" name="H+ " speciesType="t_0408" compartment="c_10" boundaryCondition="true">
Line 38218: <species id="s_0770_b" name="H+ " speciesType="t_0408" compartment="c_12" boundaryCondition="true">
Line 38225: <species id="s_0771_b" name="H+ " speciesType="t_0408" compartment="c_14" boundaryCondition="true">
Line 38232: <species id="s_0772_b" name="H+ " speciesType="t_0408" compartment="c_16" boundaryCondition="true">
Line 38239: <species id="s_2884_b" name="H+ " speciesType="t_0408" compartment="boundary" boundaryCondition="true" constant="true"/>
In my understanding, only the last instance should have boundaryCondition = "true" and the underscore b.
In line 47 of the cobra toolbox function convertSBMLtoCobra, the code looks for the underscore b's to leave those metabolites out of the model.
% Ignore exchange species
if (isempty(regexp(modelSBML.species(i).id,'_b$','ONCE')));
Again, the model runs in cobra, but it is missing many metabolites, if this is how it is supposed to happen please explain, because currently I can't make any sense of the way this is set up, and it seems to me like your SBML file is being assembled by code with a bug in it.
Ah. Yep, that's wrong, and is fixed in a newer version (maybe even 4.39 or 4.40). Silly disagreements between the SBML standard and the COBRA toolbox.
Kieran and I are in the middle of trying to finalize changes for Yeast 5 for release in the next month or two. I had asked him to hold on to the recent updates pending publication. email me at bheavner at gmail and I'll send you a more recent version, but with the warning that it's got shortcomings, too.
4.05 was the last "stable" release, so everything since that point is in active development - things are changing around a lot at the moment. We're hoping to stabilize at 5.0 in the next month or two.