You can subscribe to this list here.
| 2012 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(7) |
Sep
|
Oct
|
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2013 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(11) |
Jul
(32) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(23) |
| 2014 |
Jan
(12) |
Feb
|
Mar
(1) |
Apr
(4) |
May
(17) |
Jun
(14) |
Jul
(3) |
Aug
(26) |
Sep
(100) |
Oct
(42) |
Nov
(15) |
Dec
(6) |
| 2015 |
Jan
(3) |
Feb
|
Mar
(19) |
Apr
(4) |
May
(9) |
Jun
(4) |
Jul
(4) |
Aug
|
Sep
(2) |
Oct
(1) |
Nov
|
Dec
|
| 2016 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
|
Oct
|
Nov
(22) |
Dec
(22) |
| 2017 |
Jan
(5) |
Feb
(4) |
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
(6) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2018 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
(2) |
Oct
|
Nov
|
Dec
|
| 2019 |
Jan
(1) |
Feb
(4) |
Mar
(1) |
Apr
|
May
(1) |
Jun
|
Jul
(12) |
Aug
(2) |
Sep
|
Oct
(2) |
Nov
(6) |
Dec
(1) |
| 2020 |
Jan
|
Feb
(3) |
Mar
(1) |
Apr
|
May
(6) |
Jun
(4) |
Jul
|
Aug
|
Sep
(1) |
Oct
(1) |
Nov
|
Dec
|
| 2021 |
Jan
|
Feb
(1) |
Mar
(3) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(3) |
Nov
|
Dec
|
| 2022 |
Jan
|
Feb
|
Mar
|
Apr
(5) |
May
(1) |
Jun
|
Jul
(8) |
Aug
(3) |
Sep
|
Oct
(7) |
Nov
|
Dec
|
| S | M | T | W | T | F | S |
|---|---|---|---|---|---|---|
|
|
|
|
|
1
|
2
(4) |
3
|
|
4
|
5
|
6
(1) |
7
|
8
(1) |
9
|
10
|
|
11
|
12
|
13
(1) |
14
(1) |
15
|
16
(6) |
17
(2) |
|
18
(2) |
19
|
20
(2) |
21
|
22
|
23
(1) |
24
|
|
25
|
26
|
27
(1) |
28
|
29
|
30
|
31
|
|
From: Scott H. <snh...@nc...> - 2016-12-27 18:52:16
|
Any users of libsbml at NCSU or UNC? I am looking to discuss the sbml spatial package. -Scott |
|
From: Michael H. <mh...@ca...> - 2016-12-23 05:07:42
|
On Sun, 18 Dec 2016 12:08:21 -0500, Robert F. Murphy wrote: > I’m happy to help write the introduction if you’d like the help. I’m > attaching some review articles we have written about generating > models of cell geometries that might be useful. I’m also attaching > an extended abstract that I think was the first to actually use SBML > spatial to exchange a geometry between programs (CellOrganizer and > CellBlender), and a paper from our collaboration with Jim Faeder and > Dan Zuckerman that also uses SBML spatial to exchange geometries. Hi Bob, Thanks for your offer, and for sending those review articles. I'm currently reading these and will try to get back to this soon. Best regards, and happy holidays, MH |
|
From: Lucian S. <luc...@gm...> - 2016-12-20 23:35:42
|
A new version of the SBML Spatial specification that incorporates the few changes we've discussed recently is now available at https://sourceforge.net/p/sbml/code/HEAD/tree/trunk/specifications/sbml-level-3/version-1/spatial/specification/spatial-v1-sbml-l3v1-rel0.91.pdf The main changes are: * New examples, from Robert Murphy (truncated) * Unimplemented elements are removed - No 2D 'right triangle' CSGs - No CSG pseudo-primitives - No base64 compression * Inconsistencies made consistent - 'rotateX' not 'rotateAxisX' - 'interpolationType', not 'interpolation'. - 'roateAngleInRadians' not 'rotationAngleInRadians' * New auto-generated validation rules from Sarah Keating In addition, the SVN version of libsbml-experimental incorporates the above changes, and will be present in the next libsbml release. As always, everything can be changed more and/or reverted, depending on what people think, or what people implement; I'm hoping to make more frequent but smaller changes to the spec as we near finalization. (Mike's addition of a general introduction will probably come in release 0.92, for example.) I would still appreciate more examples that I can break down into smaller examples for the document, and I would also like to start compiling examples to share outside of the specification itself, to test exchange. Thank you all! -Lucian |
|
From: Lucian S. <luc...@gm...> - 2016-12-20 19:32:46
|
Aha! OK, that makes sense.
I have attempted to clarify the explanation in the spec, by removing the
'looking down' verbiage. Can anyone comment on whether the new text is
clear, or, if not, how you would suggest making it clearer?
I also noticed that once again, the UML and the text diverged: the UML
(and libsbml implementation) uses the name 'rotateAngleInRadians' while the
text used 'rotationAngleInRadians'. I went with 'rotate'.
---
The rotateAngleInRadians attribute}
The \token{rotateAngleInRadians} attribute is of type \primtype{double}. It
represents the rotation angle of the \CSGNode, in radians, along the
defined axis. In three-dimensional space, this rotation is defined as
counterclockwise from the perspective of the origin, viewing the
constructed vector. In two-dimensional space, this rotation is defined the
same way, as a counterclockwise rotation along the Z axis emerging from the
defined point. From the perspective of the surface of the shape, this view
looks down the negative side of the Z axis, appearing as a clockwise
rotation.
Overall, the \token{rotate} attributes define \emph{where} the rotation is
to occur, and the \token{rotateAngleInRadians} define \emph{how much}
rotation is to take place. A \token{rotateAngleInRadians} value of \val{0}
would therefore define a \CSGRotation that left the shape fixed in space.
---
I'll try to also add some clarification about the 'bottom up' way things
are interpreted.
-Lucian
On Sun, Dec 18, 2016 at 7:48 AM, Robert F. Murphy <mu...@cm...> wrote:
> Dear Lucian:
>
> Thanks for your clarification. We interpreted the rotation definition a
> third way -
>
> 1) Tilt the object to align with the vector from (0,0,0) to (rotateX,
> rotateY, rotateZ)
> 2) Rotate the object around that vector by rotateAngleInRadians.
>
> Hence you could have the rotateAngleInRadians be zero and still rotate the
> object. I think our confusion comes from interpreting the comment about
> the rotation “looking down” the vector. But now that this has been
> clarified, we can change to match what you said. I do think it would be
> helpful to clarify the description in the spec, perhaps adding an
> illustration of the resulting rotation matrix.
>
> My comment about the order of operations was about the order of applying
> the different CSGTransformations. It definitely matters - if you rotate a
> sphere before you do the scaling, no amount of rotation changes anything.
> But if you scale first and then rotate you get the desired, different
> result.
>
> So I also think the spec should be clarified to specify the order in which
> the CSGTransformations are to be applied. I think that it should be clear
> that the sequence is “bottom up” since the primitive is at the bottom but
> it would be helpful to say that.
>
>
|
|
From: Robert F. M. <mu...@cm...> - 2016-12-18 17:08:50
|
Dear Mike: I’m happy to help write the introduction if you’d like the help. I’m attaching some review articles we have written about generating models of cell geometries that might be useful. I’m also attaching an extended abstract that I think was the first to actually use SBML spatial to exchange a geometry between programs (CellOrganizer and CellBlender), and a paper from our collaboration with Jim Faeder and Dan Zuckerman that also uses SBML spatial to exchange geometries. Best, Bob ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm..., Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Dec 16, 2016, at 3:40 PM, Michael Hucka <mh...@ca...> wrote: Spatial people, In addition to what Lucian mentioned below, we could use some help in fleshing out the general introduction in the document. I know no one has time to write anything, but perhaps some of you more expert people could point us to general/basic summaries of how spatial process simulations in biology are done? My goal is to distill a roughly 1 page summary of the general principles, to give some context to all the geometric primitives and constructs used in the specification. Surely some of you have written summaries for papers or book chapters or grant proposals? (We won't take it verbatim [unless you say that's okay], just use it as source material.) MH On Thu, 15 Dec 2016 16:53:15 -0800, Lucian Smith wrote: > Hey, Jim, did you manage to get those models imported in the end? > > I'm working on a new version of the spec with the inconsistencies fixed, > and anything unimplemented removed. There's only a few things left that > I'm unsure about: > > * The 'relative complement' for CSGSets. Does VCell (or anyone else) > implement support for this? > * Quadrilateral meshes for Parametric Geometries. Does VCell (or anyone > else) implement support for this? > * CSG Homogenous Transformations: I never asked about this (just added it > to the spreadsheet), but does anyone implement reverse transformations? We > technically don't need both forward and reverse, but if anyone implemented > reverse, we can keep it. > > Things I plan to change: > * An additional requirement that coordinate Boundary objects be constant. > * Remove 2D CSGs with right triangles > * Remove CSG Pseudo-Primitives > * Remove base64 compression (unless someone wants to implement support for > it soon). > * CSGRotation attributes are now consistently 'rotateX', 'rotateY', and > 'rotateZ'. (Semantically, this could be clearer if it was changed to be > something like a child 'rotationalVector' object with X, Y, and Z > coordinates, to emphasize that these were not values to use when rotating > around the X axis, etc., but I think it's a little late for that, since > implementations already exist.) > * The SampledField attribute is now consistently 'interpolationType'. > > In addition, Sarah has managed to auto-generate a bajillion validation > rules, so those are all being added as well. Thanks, Sarah! > > -Lucian > > On Thu, Dec 1, 2016 at 10:31 PM, Schaff,Jim <sc...@uc...> wrote: > >> I imported the simple CSG, haven't tried the parametric yet. I'll get >> back to it today (Friday). >> >> >> >> ------------------------------ >> *From:* Robert F. Murphy <mu...@cm...> >> *Sent:* Wednesday, November 30, 2016 12:28 AM >> *To:* The SBML L3 Spatial Processes and Geometries package discussion list >> *Subject:* Re: [sbml-spatial] Unimplemented elements to remove from spec >> >> I believe that my group (i.e., Devin Sullivan) proposed mixed geometries >> in the last version because that is most compatible with the types of >> models learned by CellOrganizer. We really need this feature! Our >> geometries are normally generated that way (e.g., meshes for cell and >> nuclear boundaries but CSG for organelles within the cell) but for the test >> examples we generated “pure” examples to simplify testing on other >> systems. We will send some mixed examples to also test once any remaining >> issues are settled for the pure ones. Has anyone been able to successfully >> import our CSG or parametric examples? >> >> Best, >> >> Bob >> ======================= >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor >> of Biological Sciences, Biomedical Engineering, and Machine Learning >> Head, Computational Biology Department, School of Computer Science, >> Carnegie >> Mellon University >> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 >> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 >> <(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 >> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy> >> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.andrew.cmu.edu_user_murphy&d=CwMFaQ&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=IICJVVzPfNWk3T8YAcMJjEHPg1JpzXY2Gj4YGWMUXW8&s=5_G4d38QY_FAOO8Aq8dbVnchHXovZ3JLsifFz9s241Y&e= <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.andrew.cmu.edu_user_murphy&d=CwMFaQ&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=IICJVVzPfNWk3T8YAcMJjEHPg1JpzXY2Gj4YGWMUXW8&s=5_G4d38QY_FAOO8Aq8dbVnchHXovZ3JLsifFz9s241Y&e=>>), >> Twitter @murphy2537 >> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio >> Honorary Professor, University of Freiburg and External Senior Fellow, >> Freiburg Institute for Advanced Studies >> >> >> >> >> On Nov 30, 2016, at 12:02 AM, Lucian Smith <luc...@gm...> >> wrote: >> >> On Tue, Nov 29, 2016 at 7:44 PM, Schaff,Jim <sc...@uc...> wrote: >> >>> Hi Lucian, >>> >>> >>> Thanks for doing all of this. >>> >>> >>> Just filled out the survey for SBML spatial (VCell). ... a few of the >>> 'Yes" answers are not fully implemented, and none of it has been properly >>> tested with the latest SBML Spatial and latest VCell. >>> >> Thanks! That's super helpful. >> >> >> Wow, lots of options and combinations. (never considered "right triangles" >>> ... a 2D cone?). We could easily add this primitive if others find it >>> useful. >>> >> >> Nah, let's just remove it. I think I may have put it in a while back on >> the assumption that some sort of 3-sided shape would be useful, but I don't >> think it actually is, at this point. >> >>> We implement CSG pretty completely. >>> >> Excellent! You didn't fill in an answer for 'relative complement'--do you >> know if VCell implements this? >> >> >>> It seems like there are a lot of "combinations" possible with mixing and >>> matching geometric types. >>> >>> Within VCell, we used to allow mixing analytic and discrete sampled >>> fields (images) and CSG ... but not within the same domainType. >>> >>> For example, adding in an analytic (or csg) nucleus within a cell defined >>> by a sampled field (microscopy image). >>> >>> The original intent of multiple geometries was to provide >>> completely alternative representations (write many, read one). >>> >>> I answered "Yes" to combinations, VCell support was implemented so that >>> a "domainType" had to be of the same geometric representation. If >>> required >>> for interoperability with other tools, we could add support at the >>> level of >>> domains. >>> >> If I recall correctly, the idea of mixed geometries came up at a COMBINE >> or HARMONY meeting, but I don't remember which group it was that did things >> that way. It sounds like between your group and other groups, it's a >> useful enough concept to put in, though it might be a good idea to mention >> that some tools don't allow you to mix domainTypes that way, too. >> >> >>> >>> I'm agnostic about pseudo-primitives. I think that the "ordinal" concept >>> suffices and is simpler. I had proposed pseudo-primitives mainly as a for >>> formal alternative to "ordinal" order dependencies for specifying mutually >>> exclusive domains. But VCell doesn't currently use pseudo-primitives. >>> Through the "ordinal" ordering and use of "relative complements" you could >>> always "subtract" the objects with prior ordinals to remove object >>> overlap. >>> >> Well, since nobody else has implemented pseudo-primitive support either, >> it sounds like we should just get rid of it, especially if (as you say) the >> same idea can be accomplished by other means. >> >> I think that a coordinate system should be time independent. I also think >>> that object shapes/locations should be time independent also for now. >>> Motion should probably be introduced in a subsequent kinematics extension >>> (things get a lot more complicated) and remember that the geometries are >>> not meshes, they are descriptive. Motions are most generally specified by >>> velocity fields. If moving objects are strongly desired by some >>> groups, we >>> can discuss. >>> >> Dynamic spatial is probably a reasonable next step, but I think it's >> safely beyond the scope of this proposal. I didn't understand what you put >> in the spreadsheet for this: "Yes - with boundary velocity introduced'? >> Can you explain? >> >> >>> If we are going to compress the geometric representation and stick it in >>> XML, wouldn't base64 make sense? What is the alternative? Uncompressed >>> text can get pretty large for a sampled field or parametric geometry. >>> >> It would make sense, it's just not what anyone's implemented support for >> yet. People have indeed implemented support for other compression, though >> ('deflation'). >> >> You didn't say whether or not VCell implemented quadrilateral mesh >> support--do you? If not, we can simplify to just triangles, since nobody >> else has support for that, either. >> >>> I'm working through some of the example models now (and am tripping >>> over some of VCell's limitations ... e.g. a single very small "vesicle" in >>> a large span of x,y,z is missed by VCell's topology algorithm ... working >>> on it). >>> >> Thanks, that's perfect! >> >> -Lucian >> ------------------------------------------------------------ >> ------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> ------------------------------------------------------------ >> ------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, SlashDot.org! http://sdm.link/slashdot >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! > http://sdm.link/slashdot_______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org <http://slashdot.org/>! http://sdm.link/slashdot <http://sdm.link/slashdot> _______________________________________________ sbml-spatial mailing list sbm...@li... <mailto:sbm...@li...> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> |
|
From: Robert F. M. <mu...@cm...> - 2016-12-18 15:48:26
|
Dear Lucian: Thanks for your clarification. We interpreted the rotation definition a third way - 1) Tilt the object to align with the vector from (0,0,0) to (rotateX, rotateY, rotateZ) 2) Rotate the object around that vector by rotateAngleInRadians. Hence you could have the rotateAngleInRadians be zero and still rotate the object. I think our confusion comes from interpreting the comment about the rotation “looking down” the vector. But now that this has been clarified, we can change to match what you said. I do think it would be helpful to clarify the description in the spec, perhaps adding an illustration of the resulting rotation matrix. My comment about the order of operations was about the order of applying the different CSGTransformations. It definitely matters - if you rotate a sphere before you do the scaling, no amount of rotation changes anything. But if you scale first and then rotate you get the desired, different result. So I also think the spec should be clarified to specify the order in which the CSGTransformations are to be applied. I think that it should be clear that the sequence is “bottom up” since the primitive is at the bottom but it would be helpful to say that. Best, Bob ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm..., Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Dec 17, 2016, at 5:51 PM, Lucian Smith <luc...@gm...> wrote: The same issue is present with the rotations in these models as was present in the original: when spatial:rotateAngleInRadians="0", this means 'do not rotate at all'. The order doesn't matter. The rotation is supposed to happen like this: 1) Create a vector from (0,0,0) to (rotateX, rotateY, rotateZ). 2) Rotate the figure around that vector by rotateAngleInRadians. It does *not* mean 'rotate the figure around the X axis by rotateX radians, then do the same for the Y and Z axes', which is how I think you have interpreted the construct. -Lucian On Sat, Dec 17, 2016 at 11:09 AM, Ivan E Cao-Berg <ica...@an... <mailto:ica...@an...>> wrote: We generated some example synthetic images in CellOrganizer along with their SBML Spatial representation. You can find the examples here http://cellorganizer.org/SBML/examples/ <http://cellorganizer.org/SBML/examples/> The examples have been checked using the SBML online validator for syntax. > As for the rotation issue, the issue is that we probably specified the operations in the wrong order. What we want to do is start with a sphere, move it from the origin, scale it to an ellipsoid, and then tilt the ellipsoid along the line specific by the x, y, and z rotation axes. You and Jim are correct that the way we wrote it, tilting the sphere along that line doesn’t change anything. I will update the order of transformations in CellOrganizer to reflect this feedback. I hope this examples are helpful to everyone. Ivan Iván E. Cao-Berg Senior Research Programmer Computational Biology Department <http://cbd.cmu.edu/> School of Computer Science <http://www.scs.cmu.edu/> Carnegie Mellon University <http://www.cmu.edu/> <http://www.andrew.cmu.edu/~icaoberg/> <http://cbd.cmu.edu/> <http://www.twitter.com/cmucompbio> <https://www.facebook.com/cmucbd> <https://www.instagram.com/cmucbd/> <https://plus.google.com/+CMUComputationalBiologyDepartmentPittsburgh> <> <https://coderwall.com/icaoberg> <http://www.github.com/icaoberg> <http://www.bitbucket.com/icaoberg> > On Dec 16, 2016, at 2:59 PM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: > > We have some more examples that we will send shortly. > > As for the rotation issue, the issue is that we probably specified the operations in the wrong order. What we want to do is start with a sphere, move it from the origin, scale it to an ellipsoid, and then tilt the ellipsoid along the line specific by the x, y, and z rotation axes. You and Jim are correct that the way we wrote it, tilting the sphere along that line doesn’t change anything. > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Dec 16, 2016, at 1:54 PM, Lucian Smith <luc...@gm... <mailto:luc...@gm...>> wrote: > > Now that the semantic changes are in, I'm going back over these examples. Thanks again, Bob! > > I do now see what Jim pointed out: in csg_one_object, you have the following rotation: > > <spatial:csgRotation spatial:id="rotation" spatial:rotateX="2.4391" spatial:rotateY="1.5373" spatial:rotateZ="0.58404" spatial:rotateAngleInRadians="0"> > > In the definition of a csgRotation, you'll see that the X, Y, and Z values define a vector, from the origin, *around which* the modeler is to rotate the object, by the amount defined in the rotateAngleInRadians attribute. However, since in this case the rotateAngle is 0, this rotation does not actually modify the CSGObject at all. > > Is this what was intended? It seems unlikely. > > -Lucian > > On Mon, Nov 28, 2016 at 8:38 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: > resending with correct email address for Devin. > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Nov 28, 2016, at 11:34 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: > > attached are some examples we generated with CellOrganizer - they pass the online validator and roundtrip through reading and writing with libSBML. > > however, they can’t be read by VCell 5.3 (build 29). If anyone else has a program that supports SBML spatial, it would be great to see if they can be read by it. > > We will send some more examples soon, including one with mixed geometry. > > Best, > > Bob & Ivan > <SBML spatial samples CellOrganizer.zip> > > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Nov 16, 2016, at 4:24 PM, Robert F. Murphy <MU...@CM... <mailto:MU...@cm...>> wrote: > > We have generated a lot of examples using different options and are testing them using the validator and libSBML and vcell. I'm tied up today and tomorrow but will send a report and examples on Friday. > > Best, > > Bob > > Sent from my iPhone > ================ > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Carnegie Mellon University > 5000 Forbes Ave., Pittsburgh, PA 15213; +1.412.268.3480 <tel:(412)%20268-3480> > >> On Nov 16, 2016, at 1:08 PM, Frank T. Bergmann <fbe...@ca... <mailto:fbe...@ca...>> wrote: >> >> Hello Jim, >> >> as far as I'm aware the current libSBML implementation (experimental) >> implements the latest spec. If you find any issues please let me know >> and i will fix it right away. >> >> best >> Frank >> >>> On Wed, Nov 16, 2016 at 9:33 PM, Schaff,Jim <sc...@uc... <mailto:sc...@uc...>> wrote: >>> Hi All, >>> >>> Is it true that libSBML does not support the latest SBML Spatial specification? I know in theory it is orthogonal, but in practice it is essential. >>> >>> Lucian, do you or Frank know how close libSBML is to the latest spec changes? >>> >>> I'd be willing to take a whack at compliance to the latest spec, especially if I don't have to make the libSBML changes. >>> >>> Jim. >>> >>> >>> ________________________________________ >>> From: Lucian Smith <luc...@gm... <mailto:luc...@gm...>> >>> Sent: Wednesday, November 16, 2016 3:07 PM >>> To: The SBML L3 Spatial Processes and Geometries package discussion list >>> Subject: [sbml-spatial] Request: example models >>> >>> Hey, everyone! I have some time to work on the spatial specification >>> again, and it looks pretty close to being finished. At this point, >>> the main thing it needs is some examples that illustrate both the >>> commonly-used aspects of the spec, as well as a few illustrating some >>> of the potentially-confusing edge cases. From there, I think it's >>> mostly simply a matter of making sure at least one group has >>> implemented support for every element in the spec, and it'll be good >>> to go! >>> >>> If anyone has or could put together some simple spatial models that >>> follow the current specification, and could send them my way, I think >>> we can wrap this thing up and call it finished. Thank you! >>> >>> As always, if anyone has any comments on the specification itself, >>> don't hesitate to bring them up. >>> >>> -Lucian >>> >>> ------------------------------------------------------------------------------ >>> _______________________________________________ >>> sbml-spatial mailing list >>> sbm...@li... <mailto:sbm...@li...> >>> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e= <https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e=> >>> >>> ------------------------------------------------------------------------------ >>> _______________________________________________ >>> sbml-spatial mailing list >>> sbm...@li... <mailto:sbm...@li...> >>> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... <mailto:sbm...@li...> >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > sbml-spatial mailing list > sbm...@li... <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org <http://slashdot.org/>! http://sdm.link/slashdot_______________________________________________ <http://sdm.link/slashdot_______________________________________________> > sbml-spatial mailing list > sbm...@li... <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot <http://sdm.link/slashdot> _______________________________________________ sbml-spatial mailing list sbm...@li... <mailto:sbm...@li...> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot_______________________________________________ sbml-spatial mailing list sbm...@li... https://lists.sourceforge.net/lists/listinfo/sbml-spatial |
|
From: Lucian S. <luc...@gm...> - 2016-12-17 22:52:02
|
The same issue is present with the rotations in these models as was present in the original: when spatial:rotateAngleInRadians="0", this means 'do not rotate at all'. The order doesn't matter. The rotation is supposed to happen like this: 1) Create a vector from (0,0,0) to (rotateX, rotateY, rotateZ). 2) Rotate the figure around that vector by rotateAngleInRadians. It does *not* mean 'rotate the figure around the X axis by rotateX radians, then do the same for the Y and Z axes', which is how I think you have interpreted the construct. -Lucian On Sat, Dec 17, 2016 at 11:09 AM, Ivan E Cao-Berg <ica...@an...> wrote: > We generated some example synthetic images in CellOrganizer along with > their SBML Spatial representation. > > You can find the examples here > > http://cellorganizer.org/SBML/examples/ > > The examples have been checked using the SBML online validator for syntax. > > As for the rotation issue, the issue is that we probably specified the > operations in the wrong order. What we want to do is start with a sphere, > move it from the origin, scale it to an ellipsoid, and then tilt the > ellipsoid along the line specific by the x, y, and z rotation axes. You > and Jim are correct that the way we wrote it, tilting the sphere along that > line doesn’t change anything. > > > I will update the order of transformations in CellOrganizer to reflect > this feedback. > > I hope this examples are helpful to everyone. > > Ivan > > ------------------------------ > Iván E. Cao-Berg > Senior Research Programmer > Computational Biology Department <http://cbd.cmu.edu> > School of Computer Science <http://www.scs.cmu.edu> > Carnegie Mellon University <http://www.cmu.edu/> > > <http://www.andrew.cmu.edu/~icaoberg/> <http://cbd.cmu.edu> > <http://www.twitter.com/cmucompbio> <https://www.facebook.com/cmucbd> > <https://www.instagram.com/cmucbd/> > <https://plus.google.com/+CMUComputationalBiologyDepartmentPittsburgh> > <https://coderwall.com/icaoberg> <http://www.github.com/icaoberg> > <http://www.bitbucket.com/icaoberg> > > On Dec 16, 2016, at 2:59 PM, Robert F. Murphy <mu...@cm...> wrote: > > We have some more examples that we will send shortly. > > As for the rotation issue, the issue is that we probably specified the > operations in the wrong order. What we want to do is start with a sphere, > move it from the origin, scale it to an ellipsoid, and then tilt the > ellipsoid along the line specific by the x, y, and z rotation axes. You > and Jim are correct that the way we wrote it, tilting the sphere along that > line doesn’t change anything. > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor > of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie > Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 > <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter > @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, > Freiburg Institute for Advanced Studies > > > > > On Dec 16, 2016, at 1:54 PM, Lucian Smith <luc...@gm...> > wrote: > > Now that the semantic changes are in, I'm going back over these examples. > Thanks again, Bob! > > I do now see what Jim pointed out: in csg_one_object, you have the > following rotation: > > <spatial:csgRotation spatial:id="rotation" > spatial:rotateX="2.4391" spatial:rotateY="1.5373" spatial:rotateZ="0.58404" > spatial:rotateAngleInRadians="0"> > > In the definition of a csgRotation, you'll see that the X, Y, and Z values > define a vector, from the origin, *around which* the modeler is to rotate > the object, by the amount defined in the rotateAngleInRadians attribute. > However, since in this case the rotateAngle is 0, this rotation does not > actually modify the CSGObject at all. > > Is this what was intended? It seems unlikely. > > -Lucian > > On Mon, Nov 28, 2016 at 8:38 AM, Robert F. Murphy <mu...@cm...> wrote: > >> resending with correct email address for Devin. >> >> Best, >> >> Bob >> ======================= >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor >> of Biological Sciences, Biomedical Engineering, and Machine Learning >> Head, Computational Biology Department, School of Computer Science, Carnegie >> Mellon University >> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 >> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 >> <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953 >> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter >> @murphy2537 >> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio >> Honorary Professor, University of Freiburg and External Senior Fellow, >> Freiburg Institute for Advanced Studies >> >> >> >> >> On Nov 28, 2016, at 11:34 AM, Robert F. Murphy <mu...@cm...> wrote: >> >> attached are some examples we generated with CellOrganizer - they pass >> the online validator and roundtrip through reading and writing with libSBML. >> >> however, they can’t be read by VCell 5.3 (build 29). If anyone else has >> a program that supports SBML spatial, it would be great to see if they can >> be read by it. >> >> We will send some more examples soon, including one with mixed geometry. >> >> Best, >> >> Bob & Ivan >> <SBML spatial samples CellOrganizer.zip> >> >> ======================= >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor >> of Biological Sciences, Biomedical Engineering, and Machine Learning >> Head, Computational Biology Department, School of Computer Science, Carnegie >> Mellon University >> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 >> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 >> <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953 >> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter >> @murphy2537 >> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio >> Honorary Professor, University of Freiburg and External Senior Fellow, >> Freiburg Institute for Advanced Studies >> >> >> >> >> On Nov 16, 2016, at 4:24 PM, Robert F. Murphy <MU...@CM... >> <MU...@cm...>> wrote: >> >> We have generated a lot of examples using different options and are >> testing them using the validator and libSBML and vcell. I'm tied up today >> and tomorrow but will send a report and examples on Friday. >> >> Best, >> >> Bob >> >> Sent from my iPhone >> ================ >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor >> of Biological Sciences, Biomedical Engineering, and Machine Learning >> Carnegie Mellon University >> 5000 Forbes Ave., Pittsburgh, PA 15213; +1.412.268.3480 >> <(412)%20268-3480> >> >> On Nov 16, 2016, at 1:08 PM, Frank T. Bergmann <fbe...@ca...> >> wrote: >> >> Hello Jim, >> >> as far as I'm aware the current libSBML implementation (experimental) >> implements the latest spec. If you find any issues please let me know >> and i will fix it right away. >> >> best >> Frank >> >> On Wed, Nov 16, 2016 at 9:33 PM, Schaff,Jim <sc...@uc...> wrote: >> Hi All, >> >> Is it true that libSBML does not support the latest SBML Spatial >> specification? I know in theory it is orthogonal, but in practice it is >> essential. >> >> Lucian, do you or Frank know how close libSBML is to the latest spec >> changes? >> >> I'd be willing to take a whack at compliance to the latest spec, >> especially if I don't have to make the libSBML changes. >> >> Jim. >> >> >> ________________________________________ >> From: Lucian Smith <luc...@gm...> >> Sent: Wednesday, November 16, 2016 3:07 PM >> To: The SBML L3 Spatial Processes and Geometries package discussion list >> Subject: [sbml-spatial] Request: example models >> >> Hey, everyone! I have some time to work on the spatial specification >> again, and it looks pretty close to being finished. At this point, >> the main thing it needs is some examples that illustrate both the >> commonly-used aspects of the spec, as well as a few illustrating some >> of the potentially-confusing edge cases. From there, I think it's >> mostly simply a matter of making sure at least one group has >> implemented support for every element in the spec, and it'll be good >> to go! >> >> If anyone has or could put together some simple spatial models that >> follow the current specification, and could send them my way, I think >> we can wrap this thing up and call it finished. Thank you! >> >> As always, if anyone has any comments on the specification itself, >> don't hesitate to bring them up. >> >> -Lucian >> >> ------------------------------------------------------------ >> ------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.s >> ourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c= >> EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2 >> ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXX >> PSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e= >> >> ------------------------------------------------------------ >> ------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> ------------------------------------------------------------ >> ------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> >> >> >> ------------------------------------------------------------ >> ------------------ >> >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org <http://slashdot.org/>! > http://sdm.link/slashdot_______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > > > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > > |
|
From: Ivan E Cao-B. <ica...@an...> - 2016-12-17 19:33:50
|
We generated some example synthetic images in CellOrganizer along with their SBML Spatial representation. You can find the examples here http://cellorganizer.org/SBML/examples/ <http://cellorganizer.org/SBML/examples/> The examples have been checked using the SBML online validator for syntax. > As for the rotation issue, the issue is that we probably specified the operations in the wrong order. What we want to do is start with a sphere, move it from the origin, scale it to an ellipsoid, and then tilt the ellipsoid along the line specific by the x, y, and z rotation axes. You and Jim are correct that the way we wrote it, tilting the sphere along that line doesn’t change anything. I will update the order of transformations in CellOrganizer to reflect this feedback. I hope this examples are helpful to everyone. Ivan Iván E. Cao-Berg Senior Research Programmer Computational Biology Department <http://cbd.cmu.edu/> School of Computer Science <http://www.scs.cmu.edu/> Carnegie Mellon University <http://www.cmu.edu/> <http://www.andrew.cmu.edu/~icaoberg/> <http://cbd.cmu.edu/> <http://www.twitter.com/cmucompbio> <https://www.facebook.com/cmucbd> <https://www.instagram.com/cmucbd/> <https://plus.google.com/+CMUComputationalBiologyDepartmentPittsburgh> <skype:icaoberg?call> <https://coderwall.com/icaoberg> <http://www.github.com/icaoberg> <http://www.bitbucket.com/icaoberg> > On Dec 16, 2016, at 2:59 PM, Robert F. Murphy <mu...@cm...> wrote: > > We have some more examples that we will send shortly. > > As for the rotation issue, the issue is that we probably specified the operations in the wrong order. What we want to do is start with a sphere, move it from the origin, scale it to an ellipsoid, and then tilt the ellipsoid along the line specific by the x, y, and z rotation axes. You and Jim are correct that the way we wrote it, tilting the sphere along that line doesn’t change anything. > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Dec 16, 2016, at 1:54 PM, Lucian Smith <luc...@gm... <mailto:luc...@gm...>> wrote: > > Now that the semantic changes are in, I'm going back over these examples. Thanks again, Bob! > > I do now see what Jim pointed out: in csg_one_object, you have the following rotation: > > <spatial:csgRotation spatial:id="rotation" spatial:rotateX="2.4391" spatial:rotateY="1.5373" spatial:rotateZ="0.58404" spatial:rotateAngleInRadians="0"> > > In the definition of a csgRotation, you'll see that the X, Y, and Z values define a vector, from the origin, *around which* the modeler is to rotate the object, by the amount defined in the rotateAngleInRadians attribute. However, since in this case the rotateAngle is 0, this rotation does not actually modify the CSGObject at all. > > Is this what was intended? It seems unlikely. > > -Lucian > > On Mon, Nov 28, 2016 at 8:38 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: > resending with correct email address for Devin. > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Nov 28, 2016, at 11:34 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: > > attached are some examples we generated with CellOrganizer - they pass the online validator and roundtrip through reading and writing with libSBML. > > however, they can’t be read by VCell 5.3 (build 29). If anyone else has a program that supports SBML spatial, it would be great to see if they can be read by it. > > We will send some more examples soon, including one with mixed geometry. > > Best, > > Bob & Ivan > <SBML spatial samples CellOrganizer.zip> > > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies > > > > > On Nov 16, 2016, at 4:24 PM, Robert F. Murphy <MU...@CM... <mailto:MU...@cm...>> wrote: > > We have generated a lot of examples using different options and are testing them using the validator and libSBML and vcell. I'm tied up today and tomorrow but will send a report and examples on Friday. > > Best, > > Bob > > Sent from my iPhone > ================ > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning > Carnegie Mellon University > 5000 Forbes Ave., Pittsburgh, PA 15213; +1.412.268.3480 <tel:(412)%20268-3480> > >> On Nov 16, 2016, at 1:08 PM, Frank T. Bergmann <fbe...@ca... <mailto:fbe...@ca...>> wrote: >> >> Hello Jim, >> >> as far as I'm aware the current libSBML implementation (experimental) >> implements the latest spec. If you find any issues please let me know >> and i will fix it right away. >> >> best >> Frank >> >>> On Wed, Nov 16, 2016 at 9:33 PM, Schaff,Jim <sc...@uc... <mailto:sc...@uc...>> wrote: >>> Hi All, >>> >>> Is it true that libSBML does not support the latest SBML Spatial specification? I know in theory it is orthogonal, but in practice it is essential. >>> >>> Lucian, do you or Frank know how close libSBML is to the latest spec changes? >>> >>> I'd be willing to take a whack at compliance to the latest spec, especially if I don't have to make the libSBML changes. >>> >>> Jim. >>> >>> >>> ________________________________________ >>> From: Lucian Smith <luc...@gm... <mailto:luc...@gm...>> >>> Sent: Wednesday, November 16, 2016 3:07 PM >>> To: The SBML L3 Spatial Processes and Geometries package discussion list >>> Subject: [sbml-spatial] Request: example models >>> >>> Hey, everyone! I have some time to work on the spatial specification >>> again, and it looks pretty close to being finished. At this point, >>> the main thing it needs is some examples that illustrate both the >>> commonly-used aspects of the spec, as well as a few illustrating some >>> of the potentially-confusing edge cases. From there, I think it's >>> mostly simply a matter of making sure at least one group has >>> implemented support for every element in the spec, and it'll be good >>> to go! >>> >>> If anyone has or could put together some simple spatial models that >>> follow the current specification, and could send them my way, I think >>> we can wrap this thing up and call it finished. Thank you! >>> >>> As always, if anyone has any comments on the specification itself, >>> don't hesitate to bring them up. >>> >>> -Lucian >>> >>> ------------------------------------------------------------------------------ >>> _______________________________________________ >>> sbml-spatial mailing list >>> sbm...@li... <mailto:sbm...@li...> >>> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e= <https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e=> >>> >>> ------------------------------------------------------------------------------ >>> _______________________________________________ >>> sbml-spatial mailing list >>> sbm...@li... <mailto:sbm...@li...> >>> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... <mailto:sbm...@li...> >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > > > > ------------------------------------------------------------------------------ > > _______________________________________________ > sbml-spatial mailing list > sbm...@li... <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org <http://slashdot.org/>! http://sdm.link/slashdot_______________________________________________ <http://sdm.link/slashdot_______________________________________________> > sbml-spatial mailing list > sbm...@li... <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > |
|
From: Michael H. <mh...@ca...> - 2016-12-16 20:40:37
|
Spatial people, In addition to what Lucian mentioned below, we could use some help in fleshing out the general introduction in the document. I know no one has time to write anything, but perhaps some of you more expert people could point us to general/basic summaries of how spatial process simulations in biology are done? My goal is to distill a roughly 1 page summary of the general principles, to give some context to all the geometric primitives and constructs used in the specification. Surely some of you have written summaries for papers or book chapters or grant proposals? (We won't take it verbatim [unless you say that's okay], just use it as source material.) MH On Thu, 15 Dec 2016 16:53:15 -0800, Lucian Smith wrote: > Hey, Jim, did you manage to get those models imported in the end? > > I'm working on a new version of the spec with the inconsistencies fixed, > and anything unimplemented removed. There's only a few things left that > I'm unsure about: > > * The 'relative complement' for CSGSets. Does VCell (or anyone else) > implement support for this? > * Quadrilateral meshes for Parametric Geometries. Does VCell (or anyone > else) implement support for this? > * CSG Homogenous Transformations: I never asked about this (just added it > to the spreadsheet), but does anyone implement reverse transformations? We > technically don't need both forward and reverse, but if anyone implemented > reverse, we can keep it. > > Things I plan to change: > * An additional requirement that coordinate Boundary objects be constant. > * Remove 2D CSGs with right triangles > * Remove CSG Pseudo-Primitives > * Remove base64 compression (unless someone wants to implement support for > it soon). > * CSGRotation attributes are now consistently 'rotateX', 'rotateY', and > 'rotateZ'. (Semantically, this could be clearer if it was changed to be > something like a child 'rotationalVector' object with X, Y, and Z > coordinates, to emphasize that these were not values to use when rotating > around the X axis, etc., but I think it's a little late for that, since > implementations already exist.) > * The SampledField attribute is now consistently 'interpolationType'. > > In addition, Sarah has managed to auto-generate a bajillion validation > rules, so those are all being added as well. Thanks, Sarah! > > -Lucian > > On Thu, Dec 1, 2016 at 10:31 PM, Schaff,Jim <sc...@uc...> wrote: > >> I imported the simple CSG, haven't tried the parametric yet. I'll get >> back to it today (Friday). >> >> >> >> ------------------------------ >> *From:* Robert F. Murphy <mu...@cm...> >> *Sent:* Wednesday, November 30, 2016 12:28 AM >> *To:* The SBML L3 Spatial Processes and Geometries package discussion list >> *Subject:* Re: [sbml-spatial] Unimplemented elements to remove from spec >> >> I believe that my group (i.e., Devin Sullivan) proposed mixed geometries >> in the last version because that is most compatible with the types of >> models learned by CellOrganizer. We really need this feature! Our >> geometries are normally generated that way (e.g., meshes for cell and >> nuclear boundaries but CSG for organelles within the cell) but for the test >> examples we generated “pure” examples to simplify testing on other >> systems. We will send some mixed examples to also test once any remaining >> issues are settled for the pure ones. Has anyone been able to successfully >> import our CSG or parametric examples? >> >> Best, >> >> Bob >> ======================= >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor >> of Biological Sciences, Biomedical Engineering, and Machine Learning >> Head, Computational Biology Department, School of Computer Science, >> Carnegie >> Mellon University >> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 >> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 >> <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953 >> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy >> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.andrew.cmu.edu_user_murphy&d=CwMFaQ&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=IICJVVzPfNWk3T8YAcMJjEHPg1JpzXY2Gj4YGWMUXW8&s=5_G4d38QY_FAOO8Aq8dbVnchHXovZ3JLsifFz9s241Y&e=>), >> Twitter @murphy2537 >> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio >> Honorary Professor, University of Freiburg and External Senior Fellow, >> Freiburg Institute for Advanced Studies >> >> >> >> >> On Nov 30, 2016, at 12:02 AM, Lucian Smith <luc...@gm...> >> wrote: >> >> On Tue, Nov 29, 2016 at 7:44 PM, Schaff,Jim <sc...@uc...> wrote: >> >>> Hi Lucian, >>> >>> >>> Thanks for doing all of this. >>> >>> >>> Just filled out the survey for SBML spatial (VCell). ... a few of the >>> 'Yes" answers are not fully implemented, and none of it has been properly >>> tested with the latest SBML Spatial and latest VCell. >>> >> Thanks! That's super helpful. >> >> >> Wow, lots of options and combinations. (never considered "right triangles" >>> ... a 2D cone?). We could easily add this primitive if others find it >>> useful. >>> >> >> Nah, let's just remove it. I think I may have put it in a while back on >> the assumption that some sort of 3-sided shape would be useful, but I don't >> think it actually is, at this point. >> >>> We implement CSG pretty completely. >>> >> Excellent! You didn't fill in an answer for 'relative complement'--do you >> know if VCell implements this? >> >> >>> It seems like there are a lot of "combinations" possible with mixing and >>> matching geometric types. >>> >>> Within VCell, we used to allow mixing analytic and discrete sampled >>> fields (images) and CSG ... but not within the same domainType. >>> >>> For example, adding in an analytic (or csg) nucleus within a cell defined >>> by a sampled field (microscopy image). >>> >>> The original intent of multiple geometries was to provide >>> completely alternative representations (write many, read one). >>> >>> I answered "Yes" to combinations, VCell support was implemented so that >>> a "domainType" had to be of the same geometric representation. If >>> required >>> for interoperability with other tools, we could add support at the >>> level of >>> domains. >>> >> If I recall correctly, the idea of mixed geometries came up at a COMBINE >> or HARMONY meeting, but I don't remember which group it was that did things >> that way. It sounds like between your group and other groups, it's a >> useful enough concept to put in, though it might be a good idea to mention >> that some tools don't allow you to mix domainTypes that way, too. >> >> >>> >>> I'm agnostic about pseudo-primitives. I think that the "ordinal" concept >>> suffices and is simpler. I had proposed pseudo-primitives mainly as a for >>> formal alternative to "ordinal" order dependencies for specifying mutually >>> exclusive domains. But VCell doesn't currently use pseudo-primitives. >>> Through the "ordinal" ordering and use of "relative complements" you could >>> always "subtract" the objects with prior ordinals to remove object >>> overlap. >>> >> Well, since nobody else has implemented pseudo-primitive support either, >> it sounds like we should just get rid of it, especially if (as you say) the >> same idea can be accomplished by other means. >> >> I think that a coordinate system should be time independent. I also think >>> that object shapes/locations should be time independent also for now. >>> Motion should probably be introduced in a subsequent kinematics extension >>> (things get a lot more complicated) and remember that the geometries are >>> not meshes, they are descriptive. Motions are most generally specified by >>> velocity fields. If moving objects are strongly desired by some >>> groups, we >>> can discuss. >>> >> Dynamic spatial is probably a reasonable next step, but I think it's >> safely beyond the scope of this proposal. I didn't understand what you put >> in the spreadsheet for this: "Yes - with boundary velocity introduced'? >> Can you explain? >> >> >>> If we are going to compress the geometric representation and stick it in >>> XML, wouldn't base64 make sense? What is the alternative? Uncompressed >>> text can get pretty large for a sampled field or parametric geometry. >>> >> It would make sense, it's just not what anyone's implemented support for >> yet. People have indeed implemented support for other compression, though >> ('deflation'). >> >> You didn't say whether or not VCell implemented quadrilateral mesh >> support--do you? If not, we can simplify to just triangles, since nobody >> else has support for that, either. >> >>> I'm working through some of the example models now (and am tripping >>> over some of VCell's limitations ... e.g. a single very small "vesicle" in >>> a large span of x,y,z is missed by VCell's topology algorithm ... working >>> on it). >>> >> Thanks, that's perfect! >> >> -Lucian >> ------------------------------------------------------------ >> ------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> ------------------------------------------------------------ >> ------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, SlashDot.org! http://sdm.link/slashdot >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! > http://sdm.link/slashdot_______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial |
|
From: Robert F. M. <mu...@cm...> - 2016-12-16 19:59:24
|
We have some more examples that we will send shortly. As for the rotation issue, the issue is that we probably specified the operations in the wrong order. What we want to do is start with a sphere, move it from the origin, scale it to an ellipsoid, and then tilt the ellipsoid along the line specific by the x, y, and z rotation axes. You and Jim are correct that the way we wrote it, tilting the sphere along that line doesn’t change anything. Best, Bob ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm..., Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Dec 16, 2016, at 1:54 PM, Lucian Smith <luc...@gm...> wrote: Now that the semantic changes are in, I'm going back over these examples. Thanks again, Bob! I do now see what Jim pointed out: in csg_one_object, you have the following rotation: <spatial:csgRotation spatial:id="rotation" spatial:rotateX="2.4391" spatial:rotateY="1.5373" spatial:rotateZ="0.58404" spatial:rotateAngleInRadians="0"> In the definition of a csgRotation, you'll see that the X, Y, and Z values define a vector, from the origin, *around which* the modeler is to rotate the object, by the amount defined in the rotateAngleInRadians attribute. However, since in this case the rotateAngle is 0, this rotation does not actually modify the CSGObject at all. Is this what was intended? It seems unlikely. -Lucian On Mon, Nov 28, 2016 at 8:38 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: resending with correct email address for Devin. Best, Bob ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Nov 28, 2016, at 11:34 AM, Robert F. Murphy <mu...@cm... <mailto:mu...@cm...>> wrote: attached are some examples we generated with CellOrganizer - they pass the online validator and roundtrip through reading and writing with libSBML. however, they can’t be read by VCell 5.3 (build 29). If anyone else has a program that supports SBML spatial, it would be great to see if they can be read by it. We will send some more examples soon, including one with mixed geometry. Best, Bob & Ivan <SBML spatial samples CellOrganizer.zip> ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480 <tel:(412)%20268-3480>, FAX: 412-268-2977 <tel:(412)%20268-2977>, email: mu...@cm... <mailto:mu...@cm...>, Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy <http://www.andrew.cmu.edu/user/murphy>), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Nov 16, 2016, at 4:24 PM, Robert F. Murphy <MU...@CM... <mailto:MU...@cm...>> wrote: We have generated a lot of examples using different options and are testing them using the validator and libSBML and vcell. I'm tied up today and tomorrow but will send a report and examples on Friday. Best, Bob Sent from my iPhone ================ Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Carnegie Mellon University 5000 Forbes Ave., Pittsburgh, PA 15213; +1.412.268.3480 <tel:(412)%20268-3480> > On Nov 16, 2016, at 1:08 PM, Frank T. Bergmann <fbe...@ca... <mailto:fbe...@ca...>> wrote: > > Hello Jim, > > as far as I'm aware the current libSBML implementation (experimental) > implements the latest spec. If you find any issues please let me know > and i will fix it right away. > > best > Frank > >> On Wed, Nov 16, 2016 at 9:33 PM, Schaff,Jim <sc...@uc... <mailto:sc...@uc...>> wrote: >> Hi All, >> >> Is it true that libSBML does not support the latest SBML Spatial specification? I know in theory it is orthogonal, but in practice it is essential. >> >> Lucian, do you or Frank know how close libSBML is to the latest spec changes? >> >> I'd be willing to take a whack at compliance to the latest spec, especially if I don't have to make the libSBML changes. >> >> Jim. >> >> >> ________________________________________ >> From: Lucian Smith <luc...@gm... <mailto:luc...@gm...>> >> Sent: Wednesday, November 16, 2016 3:07 PM >> To: The SBML L3 Spatial Processes and Geometries package discussion list >> Subject: [sbml-spatial] Request: example models >> >> Hey, everyone! I have some time to work on the spatial specification >> again, and it looks pretty close to being finished. At this point, >> the main thing it needs is some examples that illustrate both the >> commonly-used aspects of the spec, as well as a few illustrating some >> of the potentially-confusing edge cases. From there, I think it's >> mostly simply a matter of making sure at least one group has >> implemented support for every element in the spec, and it'll be good >> to go! >> >> If anyone has or could put together some simple spatial models that >> follow the current specification, and could send them my way, I think >> we can wrap this thing up and call it finished. Thank you! >> >> As always, if anyone has any comments on the specification itself, >> don't hesitate to bring them up. >> >> -Lucian >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... <mailto:sbm...@li...> >> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e= <https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7fL67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e=> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... <mailto:sbm...@li...> >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > ------------------------------------------------------------------------------ > _______________________________________________ > sbml-spatial mailing list > sbm...@li... <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> ------------------------------------------------------------------------------ _______________________________________________ sbml-spatial mailing list sbm...@li... <mailto:sbm...@li...> https://lists.sourceforge.net/lists/listinfo/sbml-spatial <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot_______________________________________________ sbml-spatial mailing list sbm...@li... https://lists.sourceforge.net/lists/listinfo/sbml-spatial |
|
From: Lucian S. <luc...@gm...> - 2016-12-16 18:54:11
|
Now that the semantic changes are in, I'm going back over these examples.
Thanks again, Bob!
I do now see what Jim pointed out: in csg_one_object, you have the
following rotation:
<spatial:csgRotation spatial:id="rotation"
spatial:rotateX="2.4391" spatial:rotateY="1.5373" spatial:rotateZ="0.58404"
spatial:rotateAngleInRadians="0">
In the definition of a csgRotation, you'll see that the X, Y, and Z values
define a vector, from the origin, *around which* the modeler is to rotate
the object, by the amount defined in the rotateAngleInRadians attribute.
However, since in this case the rotateAngle is 0, this rotation does not
actually modify the CSGObject at all.
Is this what was intended? It seems unlikely.
-Lucian
On Mon, Nov 28, 2016 at 8:38 AM, Robert F. Murphy <mu...@cm...> wrote:
> resending with correct email address for Devin.
>
> Best,
>
> Bob
> =======================
> Robert F. Murphy, Ph.D.
> Ray and Stephanie Lane Professor of Computational Biology and Professor
> of Biological Sciences, Biomedical Engineering, and Machine Learning
> Head, Computational Biology Department, School of Computer Science, Carnegie
> Mellon University
> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213
> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977
> <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953
> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter
> @murphy2537
> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio
> Honorary Professor, University of Freiburg and External Senior Fellow,
> Freiburg Institute for Advanced Studies
>
>
>
>
> On Nov 28, 2016, at 11:34 AM, Robert F. Murphy <mu...@cm...> wrote:
>
> attached are some examples we generated with CellOrganizer - they pass the
> online validator and roundtrip through reading and writing with libSBML.
>
> however, they can’t be read by VCell 5.3 (build 29). If anyone else has a
> program that supports SBML spatial, it would be great to see if they can be
> read by it.
>
> We will send some more examples soon, including one with mixed geometry.
>
> Best,
>
> Bob & Ivan
> <SBML spatial samples CellOrganizer.zip>
>
> =======================
> Robert F. Murphy, Ph.D.
> Ray and Stephanie Lane Professor of Computational Biology and Professor
> of Biological Sciences, Biomedical Engineering, and Machine Learning
> Head, Computational Biology Department, School of Computer Science, Carnegie
> Mellon University
> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213
> Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977
> <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953
> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter
> @murphy2537
> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio
> Honorary Professor, University of Freiburg and External Senior Fellow,
> Freiburg Institute for Advanced Studies
>
>
>
>
> On Nov 16, 2016, at 4:24 PM, Robert F. Murphy <MU...@CM...
> <MU...@cm...>> wrote:
>
> We have generated a lot of examples using different options and are
> testing them using the validator and libSBML and vcell. I'm tied up today
> and tomorrow but will send a report and examples on Friday.
>
> Best,
>
> Bob
>
> Sent from my iPhone
> ================
> Robert F. Murphy, Ph.D.
> Ray and Stephanie Lane Professor of Computational Biology and Professor of
> Biological Sciences, Biomedical Engineering, and Machine Learning
> Carnegie Mellon University
> 5000 Forbes Ave., Pittsburgh, PA 15213; +1.412.268.3480 <(412)%20268-3480>
>
> On Nov 16, 2016, at 1:08 PM, Frank T. Bergmann <fbe...@ca...>
> wrote:
>
> Hello Jim,
>
> as far as I'm aware the current libSBML implementation (experimental)
> implements the latest spec. If you find any issues please let me know
> and i will fix it right away.
>
> best
> Frank
>
> On Wed, Nov 16, 2016 at 9:33 PM, Schaff,Jim <sc...@uc...> wrote:
> Hi All,
>
> Is it true that libSBML does not support the latest SBML Spatial
> specification? I know in theory it is orthogonal, but in practice it is
> essential.
>
> Lucian, do you or Frank know how close libSBML is to the latest spec
> changes?
>
> I'd be willing to take a whack at compliance to the latest spec,
> especially if I don't have to make the libSBML changes.
>
> Jim.
>
>
> ________________________________________
> From: Lucian Smith <luc...@gm...>
> Sent: Wednesday, November 16, 2016 3:07 PM
> To: The SBML L3 Spatial Processes and Geometries package discussion list
> Subject: [sbml-spatial] Request: example models
>
> Hey, everyone! I have some time to work on the spatial specification
> again, and it looks pretty close to being finished. At this point,
> the main thing it needs is some examples that illustrate both the
> commonly-used aspects of the spec, as well as a few illustrating some
> of the potentially-confusing edge cases. From there, I think it's
> mostly simply a matter of making sure at least one group has
> implemented support for every element in the spec, and it'll be good
> to go!
>
> If anyone has or could put together some simple spatial models that
> follow the current specification, and could send them my way, I think
> we can wrap this thing up and call it finished. Thank you!
>
> As always, if anyone has any comments on the specification itself,
> don't hesitate to bring them up.
>
> -Lucian
>
> ------------------------------------------------------------
> ------------------
> _______________________________________________
> sbml-spatial mailing list
> sbm...@li...
> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.
> sourceforge.net_lists_listinfo_sbml-2Dspatial&d=CwICAg&c=EZxp_
> D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1
> CI2zO48jPRDC4&m=D71-76AqMfX4FPnO8tmiPyeCvKXXPSbv7f
> L67OdPOjk&s=zQwyhcdtfpS2YhK0q_zjdryBvhK6vNBFtf_7mAYHpdo&e=
>
> ------------------------------------------------------------
> ------------------
> _______________________________________________
> sbml-spatial mailing list
> sbm...@li...
> https://lists.sourceforge.net/lists/listinfo/sbml-spatial
>
>
> ------------------------------------------------------------
> ------------------
> _______________________________________________
> sbml-spatial mailing list
> sbm...@li...
> https://lists.sourceforge.net/lists/listinfo/sbml-spatial
>
>
>
>
>
> ------------------------------------------------------------
> ------------------
>
> _______________________________________________
> sbml-spatial mailing list
> sbm...@li...
> https://lists.sourceforge.net/lists/listinfo/sbml-spatial
>
>
|
|
From: Sarah K. <ske...@ca...> - 2016-12-16 10:57:17
|
> In addition, Sarah has managed to auto-generate a bajillion validation > rules, so those are all being added as well. Thanks, Sarah! > Remember to let me know so I can regenerate these once you have removed/changed anything you are going to change. It would be better to let Deviser do this so numbers in code remain consistent with numbers in the spec. Sarah |
|
From: Frank T. B. <fbe...@ca...> - 2016-12-16 08:03:13
|
Hello Lucian thanks for doing that, that plan looks just fine to me ... best Frank On Fri, Dec 16, 2016 at 1:53 AM, Lucian Smith <luc...@gm...> wrote: > Hey, Jim, did you manage to get those models imported in the end? > > I'm working on a new version of the spec with the inconsistencies fixed, and > anything unimplemented removed. There's only a few things left that I'm > unsure about: > > * The 'relative complement' for CSGSets. Does VCell (or anyone else) > implement support for this? > * Quadrilateral meshes for Parametric Geometries. Does VCell (or anyone > else) implement support for this? > * CSG Homogenous Transformations: I never asked about this (just added it > to the spreadsheet), but does anyone implement reverse transformations? We > technically don't need both forward and reverse, but if anyone implemented > reverse, we can keep it. > > Things I plan to change: > * An additional requirement that coordinate Boundary objects be constant. > * Remove 2D CSGs with right triangles > * Remove CSG Pseudo-Primitives > * Remove base64 compression (unless someone wants to implement support for > it soon). > * CSGRotation attributes are now consistently 'rotateX', 'rotateY', and > 'rotateZ'. (Semantically, this could be clearer if it was changed to be > something like a child 'rotationalVector' object with X, Y, and Z > coordinates, to emphasize that these were not values to use when rotating > around the X axis, etc., but I think it's a little late for that, since > implementations already exist.) > * The SampledField attribute is now consistently 'interpolationType'. > > In addition, Sarah has managed to auto-generate a bajillion validation > rules, so those are all being added as well. Thanks, Sarah! > > -Lucian > > On Thu, Dec 1, 2016 at 10:31 PM, Schaff,Jim <sc...@uc...> wrote: >> >> I imported the simple CSG, haven't tried the parametric yet. I'll get >> back to it today (Friday). >> >> >> >> ________________________________ >> From: Robert F. Murphy <mu...@cm...> >> Sent: Wednesday, November 30, 2016 12:28 AM >> To: The SBML L3 Spatial Processes and Geometries package discussion list >> Subject: Re: [sbml-spatial] Unimplemented elements to remove from spec >> >> I believe that my group (i.e., Devin Sullivan) proposed mixed geometries >> in the last version because that is most compatible with the types of models >> learned by CellOrganizer. We really need this feature! Our geometries are >> normally generated that way (e.g., meshes for cell and nuclear boundaries >> but CSG for organelles within the cell) but for the test examples we >> generated “pure” examples to simplify testing on other systems. We will >> send some mixed examples to also test once any remaining issues are settled >> for the pure ones. Has anyone been able to successfully import our CSG or >> parametric examples? >> >> Best, >> >> Bob >> ======================= >> Robert F. Murphy, Ph.D. >> Ray and Stephanie Lane Professor of Computational Biology and Professor of >> Biological Sciences, Biomedical Engineering, and Machine Learning >> Head, Computational Biology Department, School of Computer Science, >> Carnegie Mellon University >> 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 >> Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm..., Skype: >> rfmurphy1953 >> Personal: Home Page (http://www.andrew.cmu.edu/user/murphy), Twitter >> @murphy2537 >> Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio >> Honorary Professor, University of Freiburg and External Senior Fellow, >> Freiburg Institute for Advanced Studies >> >> >> >> >> On Nov 30, 2016, at 12:02 AM, Lucian Smith <luc...@gm...> >> wrote: >> >> On Tue, Nov 29, 2016 at 7:44 PM, Schaff,Jim <sc...@uc...> wrote: >>> >>> Hi Lucian, >>> >>> >>> Thanks for doing all of this. >>> >>> >>> Just filled out the survey for SBML spatial (VCell). ... a few of the >>> 'Yes" answers are not fully implemented, and none of it has been properly >>> tested with the latest SBML Spatial and latest VCell. >> >> Thanks! That's super helpful. >> >> >>> Wow, lots of options and combinations. (never considered "right >>> triangles" ... a 2D cone?). We could easily add this primitive if others >>> find it useful. >> >> >> Nah, let's just remove it. I think I may have put it in a while back on >> the assumption that some sort of 3-sided shape would be useful, but I don't >> think it actually is, at this point. >>> >>> We implement CSG pretty completely. >> >> Excellent! You didn't fill in an answer for 'relative complement'--do you >> know if VCell implements this? >> >>> >>> It seems like there are a lot of "combinations" possible with mixing and >>> matching geometric types. >>> >>> Within VCell, we used to allow mixing analytic and discrete sampled >>> fields (images) and CSG ... but not within the same domainType. >>> >>> For example, adding in an analytic (or csg) nucleus within a cell defined >>> by a sampled field (microscopy image). >>> >>> The original intent of multiple geometries was to provide completely >>> alternative representations (write many, read one). >>> >>> I answered "Yes" to combinations, VCell support was implemented so that a >>> "domainType" had to be of the same geometric representation. If required >>> for interoperability with other tools, we could add support at the level of >>> domains. >> >> If I recall correctly, the idea of mixed geometries came up at a COMBINE >> or HARMONY meeting, but I don't remember which group it was that did things >> that way. It sounds like between your group and other groups, it's a useful >> enough concept to put in, though it might be a good idea to mention that >> some tools don't allow you to mix domainTypes that way, too. >> >>> >>> >>> I'm agnostic about pseudo-primitives. I think that the "ordinal" concept >>> suffices and is simpler. I had proposed pseudo-primitives mainly as a for >>> formal alternative to "ordinal" order dependencies for specifying mutually >>> exclusive domains. But VCell doesn't currently use pseudo-primitives. >>> Through the "ordinal" ordering and use of "relative complements" you could >>> always "subtract" the objects with prior ordinals to remove object overlap. >> >> Well, since nobody else has implemented pseudo-primitive support either, >> it sounds like we should just get rid of it, especially if (as you say) the >> same idea can be accomplished by other means. >> >>> I think that a coordinate system should be time independent. I also >>> think that object shapes/locations should be time independent also for now. >>> Motion should probably be introduced in a subsequent kinematics extension >>> (things get a lot more complicated) and remember that the geometries are not >>> meshes, they are descriptive. Motions are most generally specified by >>> velocity fields. If moving objects are strongly desired by some groups, we >>> can discuss. >> >> Dynamic spatial is probably a reasonable next step, but I think it's >> safely beyond the scope of this proposal. I didn't understand what you put >> in the spreadsheet for this: "Yes - with boundary velocity introduced'? >> Can you explain? >> >>> >>> If we are going to compress the geometric representation and stick it in >>> XML, wouldn't base64 make sense? What is the alternative? Uncompressed >>> text can get pretty large for a sampled field or parametric geometry. >> >> It would make sense, it's just not what anyone's implemented support for >> yet. People have indeed implemented support for other compression, though >> ('deflation'). >> >> You didn't say whether or not VCell implemented quadrilateral mesh >> support--do you? If not, we can simplify to just triangles, since nobody >> else has support for that, either. >>> >>> I'm working through some of the example models now (and am tripping over >>> some of VCell's limitations ... e.g. a single very small "vesicle" in a >>> large span of x,y,z is missed by VCell's topology algorithm ... working on >>> it). >> >> Thanks, that's perfect! >> >> -Lucian >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> >> ------------------------------------------------------------------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, SlashDot.org! http://sdm.link/slashdot >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > |
|
From: Lucian S. <luc...@gm...> - 2016-12-16 00:53:24
|
Hey, Jim, did you manage to get those models imported in the end? I'm working on a new version of the spec with the inconsistencies fixed, and anything unimplemented removed. There's only a few things left that I'm unsure about: * The 'relative complement' for CSGSets. Does VCell (or anyone else) implement support for this? * Quadrilateral meshes for Parametric Geometries. Does VCell (or anyone else) implement support for this? * CSG Homogenous Transformations: I never asked about this (just added it to the spreadsheet), but does anyone implement reverse transformations? We technically don't need both forward and reverse, but if anyone implemented reverse, we can keep it. Things I plan to change: * An additional requirement that coordinate Boundary objects be constant. * Remove 2D CSGs with right triangles * Remove CSG Pseudo-Primitives * Remove base64 compression (unless someone wants to implement support for it soon). * CSGRotation attributes are now consistently 'rotateX', 'rotateY', and 'rotateZ'. (Semantically, this could be clearer if it was changed to be something like a child 'rotationalVector' object with X, Y, and Z coordinates, to emphasize that these were not values to use when rotating around the X axis, etc., but I think it's a little late for that, since implementations already exist.) * The SampledField attribute is now consistently 'interpolationType'. In addition, Sarah has managed to auto-generate a bajillion validation rules, so those are all being added as well. Thanks, Sarah! -Lucian On Thu, Dec 1, 2016 at 10:31 PM, Schaff,Jim <sc...@uc...> wrote: > I imported the simple CSG, haven't tried the parametric yet. I'll get > back to it today (Friday). > > > > ------------------------------ > *From:* Robert F. Murphy <mu...@cm...> > *Sent:* Wednesday, November 30, 2016 12:28 AM > *To:* The SBML L3 Spatial Processes and Geometries package discussion list > *Subject:* Re: [sbml-spatial] Unimplemented elements to remove from spec > > I believe that my group (i.e., Devin Sullivan) proposed mixed geometries > in the last version because that is most compatible with the types of > models learned by CellOrganizer. We really need this feature! Our > geometries are normally generated that way (e.g., meshes for cell and > nuclear boundaries but CSG for organelles within the cell) but for the test > examples we generated “pure” examples to simplify testing on other > systems. We will send some mixed examples to also test once any remaining > issues are settled for the pure ones. Has anyone been able to successfully > import our CSG or parametric examples? > > Best, > > Bob > ======================= > Robert F. Murphy, Ph.D. > Ray and Stephanie Lane Professor of Computational Biology and Professor > of Biological Sciences, Biomedical Engineering, and Machine Learning > Head, Computational Biology Department, School of Computer Science, Carnegie > Mellon University > 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 > Phone: 412-268-3480 <(412)%20268-3480>, FAX: 412-268-2977 > <(412)%20268-2977>, email: mu...@cm..., Skype: rfmurphy1953 > Personal: Home Page (http://www.andrew.cmu.edu/user/murphy > <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.andrew.cmu.edu_user_murphy&d=CwMFaQ&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=IICJVVzPfNWk3T8YAcMJjEHPg1JpzXY2Gj4YGWMUXW8&s=5_G4d38QY_FAOO8Aq8dbVnchHXovZ3JLsifFz9s241Y&e=>), > Twitter @murphy2537 > Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio > Honorary Professor, University of Freiburg and External Senior Fellow, > Freiburg Institute for Advanced Studies > > > > > On Nov 30, 2016, at 12:02 AM, Lucian Smith <luc...@gm...> > wrote: > > On Tue, Nov 29, 2016 at 7:44 PM, Schaff,Jim <sc...@uc...> wrote: > >> Hi Lucian, >> >> >> Thanks for doing all of this. >> >> >> Just filled out the survey for SBML spatial (VCell). ... a few of the >> 'Yes" answers are not fully implemented, and none of it has been properly >> tested with the latest SBML Spatial and latest VCell. >> > Thanks! That's super helpful. > > > Wow, lots of options and combinations. (never considered "right triangles" >> ... a 2D cone?). We could easily add this primitive if others find it >> useful. >> > > Nah, let's just remove it. I think I may have put it in a while back on > the assumption that some sort of 3-sided shape would be useful, but I don't > think it actually is, at this point. > >> We implement CSG pretty completely. >> > Excellent! You didn't fill in an answer for 'relative complement'--do you > know if VCell implements this? > > >> It seems like there are a lot of "combinations" possible with mixing and >> matching geometric types. >> >> Within VCell, we used to allow mixing analytic and discrete sampled >> fields (images) and CSG ... but not within the same domainType. >> >> For example, adding in an analytic (or csg) nucleus within a cell defined >> by a sampled field (microscopy image). >> >> The original intent of multiple geometries was to provide >> completely alternative representations (write many, read one). >> >> I answered "Yes" to combinations, VCell support was implemented so that >> a "domainType" had to be of the same geometric representation. If required >> for interoperability with other tools, we could add support at the level of >> domains. >> > If I recall correctly, the idea of mixed geometries came up at a COMBINE > or HARMONY meeting, but I don't remember which group it was that did things > that way. It sounds like between your group and other groups, it's a > useful enough concept to put in, though it might be a good idea to mention > that some tools don't allow you to mix domainTypes that way, too. > > >> >> I'm agnostic about pseudo-primitives. I think that the "ordinal" concept >> suffices and is simpler. I had proposed pseudo-primitives mainly as a for >> formal alternative to "ordinal" order dependencies for specifying mutually >> exclusive domains. But VCell doesn't currently use pseudo-primitives. >> Through the "ordinal" ordering and use of "relative complements" you could >> always "subtract" the objects with prior ordinals to remove object overlap. >> > Well, since nobody else has implemented pseudo-primitive support either, > it sounds like we should just get rid of it, especially if (as you say) the > same idea can be accomplished by other means. > > I think that a coordinate system should be time independent. I also think >> that object shapes/locations should be time independent also for now. >> Motion should probably be introduced in a subsequent kinematics extension >> (things get a lot more complicated) and remember that the geometries are >> not meshes, they are descriptive. Motions are most generally specified by >> velocity fields. If moving objects are strongly desired by some groups, we >> can discuss. >> > Dynamic spatial is probably a reasonable next step, but I think it's > safely beyond the scope of this proposal. I didn't understand what you put > in the spreadsheet for this: "Yes - with boundary velocity introduced'? > Can you explain? > > >> If we are going to compress the geometric representation and stick it in >> XML, wouldn't base64 make sense? What is the alternative? Uncompressed >> text can get pretty large for a sampled field or parametric geometry. >> > It would make sense, it's just not what anyone's implemented support for > yet. People have indeed implemented support for other compression, though > ('deflation'). > > You didn't say whether or not VCell implemented quadrilateral mesh > support--do you? If not, we can simplify to just triangles, since nobody > else has support for that, either. > >> I'm working through some of the example models now (and am tripping >> over some of VCell's limitations ... e.g. a single very small "vesicle" in >> a large span of x,y,z is missed by VCell's topology algorithm ... working >> on it). >> > Thanks, that's perfect! > > -Lucian > ------------------------------------------------------------ > ------------------ > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > > |
|
From: Frank T. B. <fbe...@ca...> - 2016-12-14 07:16:03
|
That works for me, the important thing is that the two are the same, and i also have a slight preference for `interpolationType`. thanks Frank On Tue, Dec 13, 2016 at 10:47 PM, Lucian Smith <luc...@gm...> wrote: > OK, if nobody else objects, then, let's go with 'interpolationType' as > canonical, and not the plain 'interpolation', as the attribute for > SampledField. > > -Lucian > > On Thu, Dec 8, 2016 at 8:58 AM, Nicolas Rodriguez <rod...@eb...> > wrote: >> >> On 06/12/16 19:08, Lucian Smith wrote: >> >> I don't think there's any semantic reason to choose one over the other, >> which means that it would probably be best to choose based on the path of >> least inconvenience. For people implementing support for SampledField, are >> there more of you using libsbml or jsbml? >> >> >> I think the changes for either libsbml or jsbml are minimal and we can >> keep the current API so that users >> of both libraries don't necessary need to change anything. So I think you >> can just pick one Lucian >> as nobody seem to be very attached to one or the other attribute name. >> >> Nico >> >> >> >> On Fri, Dec 2, 2016 at 8:11 AM, Nicolas Rodriguez <rod...@eb...> >> wrote: >>> >>> On 02/12/16 15:32, Akira Funahashi wrote: >>> > Hi, >>> > >>> > We wish if libSBML and JSBML both uses same term. >>> > During our implementation we found that >>> > libSBML exported as "interpolation", and >>> > JSBML exported as "interpolationType" for the same attribute. >>> >>> I would even say, they should use the same name :-) >>> >>> Once it is decided which name should be used, we can update what need to >>> be. >>> >>> Nico >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> Check out the vibrant tech community on one of the world's most >>> engaging tech sites, SlashDot.org! http://sdm.link/slashdot >>> _______________________________________________ >>> sbml-spatial mailing list >>> sbm...@li... >>> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> >> >> >> ------------------------------------------------------------------------------ >> Developer Access Program for Intel Xeon Phi Processors >> Access to Intel Xeon Phi processor-based developer platforms. >> With one year of Intel Parallel Studio XE. >> Training and support from Colfax. >> Order your platform today.http://sdm.link/xeonphi >> >> >> >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> >> >> >> >> ------------------------------------------------------------------------------ >> Developer Access Program for Intel Xeon Phi Processors >> Access to Intel Xeon Phi processor-based developer platforms. >> With one year of Intel Parallel Studio XE. >> Training and support from Colfax. >> Order your platform today.http://sdm.link/xeonphi >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > |
|
From: Lucian S. <luc...@gm...> - 2016-12-13 21:47:41
|
OK, if nobody else objects, then, let's go with 'interpolationType' as canonical, and not the plain 'interpolation', as the attribute for SampledField. -Lucian On Thu, Dec 8, 2016 at 8:58 AM, Nicolas Rodriguez <rod...@eb...> wrote: > On 06/12/16 19:08, Lucian Smith wrote: > > I don't think there's any semantic reason to choose one over the other, > which means that it would probably be best to choose based on the path of > least inconvenience. For people implementing support for SampledField, are > there more of you using libsbml or jsbml? > > > I think the changes for either libsbml or jsbml are minimal and we can > keep the current API so that users > of both libraries don't necessary need to change anything. So I think you > can just pick one Lucian > as nobody seem to be very attached to one or the other attribute name. > > Nico > > > > On Fri, Dec 2, 2016 at 8:11 AM, Nicolas Rodriguez <rod...@eb...> > wrote: > >> On 02/12/16 15:32, Akira Funahashi wrote: >> > Hi, >> > >> > We wish if libSBML and JSBML both uses same term. >> > During our implementation we found that >> > libSBML exported as "interpolation", and >> > JSBML exported as "interpolationType" for the same attribute. >> >> I would even say, they should use the same name :-) >> >> Once it is decided which name should be used, we can update what need to >> be. >> >> Nico >> >> >> ------------------------------------------------------------ >> ------------------ >> Check out the vibrant tech community on one of the world's most >> engaging tech sites, SlashDot.org! http://sdm.link/slashdot >> _______________________________________________ >> sbml-spatial mailing list >> sbm...@li... >> https://lists.sourceforge.net/lists/listinfo/sbml-spatial >> > > > > ------------------------------------------------------------------------------ > Developer Access Program for Intel Xeon Phi Processors > Access to Intel Xeon Phi processor-based developer platforms. > With one year of Intel Parallel Studio XE. > Training and support from Colfax. > Order your platform today.http://sdm.link/xeonphi > > > > _______________________________________________ > sbml-spatial mailing lis...@li...://lists.sourceforge.net/lists/listinfo/sbml-spatial > > > > ------------------------------------------------------------ > ------------------ > Developer Access Program for Intel Xeon Phi Processors > Access to Intel Xeon Phi processor-based developer platforms. > With one year of Intel Parallel Studio XE. > Training and support from Colfax. > Order your platform today.http://sdm.link/xeonphi > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > > |
|
From: Nicolas R. <rod...@eb...> - 2016-12-08 17:00:19
|
On 06/12/16 19:08, Lucian Smith wrote: > I don't think there's any semantic reason to choose one over the > other, which means that it would probably be best to choose based on > the path of least inconvenience. For people implementing support for > SampledField, are there more of you using libsbml or jsbml? > I think the changes for either libsbml or jsbml are minimal and we can keep the current API so that users of both libraries don't necessary need to change anything. So I think you can just pick one Lucian as nobody seem to be very attached to one or the other attribute name. Nico > > > On Fri, Dec 2, 2016 at 8:11 AM, Nicolas Rodriguez <rod...@eb... > <mailto:rod...@eb...>> wrote: > > On 02/12/16 15:32, Akira Funahashi wrote: > > Hi, > > > > We wish if libSBML and JSBML both uses same term. > > During our implementation we found that > > libSBML exported as "interpolation", and > > JSBML exported as "interpolationType" for the same attribute. > > I would even say, they should use the same name :-) > > Once it is decided which name should be used, we can update what > need to be. > > Nico > > > ------------------------------------------------------------------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > <mailto:sbm...@li...> > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > <https://lists.sourceforge.net/lists/listinfo/sbml-spatial> > > > > > ------------------------------------------------------------------------------ > Developer Access Program for Intel Xeon Phi Processors > Access to Intel Xeon Phi processor-based developer platforms. > With one year of Intel Parallel Studio XE. > Training and support from Colfax. > Order your platform today.http://sdm.link/xeonphi > > > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial |
|
From: Lucian S. <luc...@gm...> - 2016-12-06 19:08:44
|
I don't think there's any semantic reason to choose one over the other, which means that it would probably be best to choose based on the path of least inconvenience. For people implementing support for SampledField, are there more of you using libsbml or jsbml? -Lucian On Fri, Dec 2, 2016 at 8:11 AM, Nicolas Rodriguez <rod...@eb...> wrote: > On 02/12/16 15:32, Akira Funahashi wrote: > > Hi, > > > > We wish if libSBML and JSBML both uses same term. > > During our implementation we found that > > libSBML exported as "interpolation", and > > JSBML exported as "interpolationType" for the same attribute. > > I would even say, they should use the same name :-) > > Once it is decided which name should be used, we can update what need to > be. > > Nico > > > ------------------------------------------------------------ > ------------------ > Check out the vibrant tech community on one of the world's most > engaging tech sites, SlashDot.org! http://sdm.link/slashdot > _______________________________________________ > sbml-spatial mailing list > sbm...@li... > https://lists.sourceforge.net/lists/listinfo/sbml-spatial > |
|
From: Nicolas R. <rod...@eb...> - 2016-12-02 16:13:04
|
On 02/12/16 15:32, Akira Funahashi wrote: > Hi, > > We wish if libSBML and JSBML both uses same term. > During our implementation we found that > libSBML exported as "interpolation", and > JSBML exported as "interpolationType" for the same attribute. I would even say, they should use the same name :-) Once it is decided which name should be used, we can update what need to be. Nico |
|
From: Akira F. <fu...@bi...> - 2016-12-02 15:33:51
|
Hi, We wish if libSBML and JSBML both uses same term. During our implementation we found that libSBML exported as "interpolation", and JSBML exported as "interpolationType" for the same attribute. Best, -- Akira |
|
From: Sarah K. <ske...@ca...> - 2016-12-02 12:01:46
|
Hi Guys There is another mismatch in the attribute name interpolationType on a SampledField element; text (and libsbml) use 'interpolationType' but the UML diagram (and RNG schema) use 'interpolation'. Which shall we go with ? Sarah |
|
From: Schaff,Jim <sc...@uc...> - 2016-12-02 06:32:02
|
I imported the simple CSG, haven't tried the parametric yet. I'll get back to it today (Friday). ________________________________ From: Robert F. Murphy <mu...@cm...> Sent: Wednesday, November 30, 2016 12:28 AM To: The SBML L3 Spatial Processes and Geometries package discussion list Subject: Re: [sbml-spatial] Unimplemented elements to remove from spec I believe that my group (i.e., Devin Sullivan) proposed mixed geometries in the last version because that is most compatible with the types of models learned by CellOrganizer. We really need this feature! Our geometries are normally generated that way (e.g., meshes for cell and nuclear boundaries but CSG for organelles within the cell) but for the test examples we generated “pure” examples to simplify testing on other systems. We will send some mixed examples to also test once any remaining issues are settled for the pure ones. Has anyone been able to successfully import our CSG or parametric examples? Best, Bob ======================= Robert F. Murphy, Ph.D. Ray and Stephanie Lane Professor of Computational Biology and Professor of Biological Sciences, Biomedical Engineering, and Machine Learning Head, Computational Biology Department, School of Computer Science, Carnegie Mellon University 7723 Gates and Hillman Centers, 5000 Forbes Ave., Pittsburgh, PA 15213 Phone: 412-268-3480, FAX: 412-268-2977, email: mu...@cm...<mailto:mu...@cm...>, Skype: rfmurphy1953 Personal: Home Page (http://www.andrew.cmu.edu/user/murphy<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.andrew.cmu.edu_user_murphy&d=CwMFaQ&c=EZxp_D7cDnouwj5YEFHgXuSKoUq2zVQZ_7Fw9yfotck&r=6W0wBewzLZ4dT2ku5JQeMaL3BGQ7J1CI2zO48jPRDC4&m=IICJVVzPfNWk3T8YAcMJjEHPg1JpzXY2Gj4YGWMUXW8&s=5_G4d38QY_FAOO8Aq8dbVnchHXovZ3JLsifFz9s241Y&e=>), Twitter @murphy2537 Comp Bio Department: Facebook (cmucbd), Twitter @CMUCompBio Honorary Professor, University of Freiburg and External Senior Fellow, Freiburg Institute for Advanced Studies On Nov 30, 2016, at 12:02 AM, Lucian Smith <luc...@gm...<mailto:luc...@gm...>> wrote: On Tue, Nov 29, 2016 at 7:44 PM, Schaff,Jim <sc...@uc...<mailto:sc...@uc...>> wrote: Hi Lucian, Thanks for doing all of this. Just filled out the survey for SBML spatial (VCell). ... a few of the 'Yes" answers are not fully implemented, and none of it has been properly tested with the latest SBML Spatial and latest VCell. Thanks! That's super helpful. Wow, lots of options and combinations. (never considered "right triangles" ... a 2D cone?). We could easily add this primitive if others find it useful. Nah, let's just remove it. I think I may have put it in a while back on the assumption that some sort of 3-sided shape would be useful, but I don't think it actually is, at this point. We implement CSG pretty completely. Excellent! You didn't fill in an answer for 'relative complement'--do you know if VCell implements this? It seems like there are a lot of "combinations" possible with mixing and matching geometric types. Within VCell, we used to allow mixing analytic and discrete sampled fields (images) and CSG ... but not within the same domainType. For example, adding in an analytic (or csg) nucleus within a cell defined by a sampled field (microscopy image). The original intent of multiple geometries was to provide completely alternative representations (write many, read one). I answered "Yes" to combinations, VCell support was implemented so that a "domainType" had to be of the same geometric representation. If required for interoperability with other tools, we could add support at the level of domains. If I recall correctly, the idea of mixed geometries came up at a COMBINE or HARMONY meeting, but I don't remember which group it was that did things that way. It sounds like between your group and other groups, it's a useful enough concept to put in, though it might be a good idea to mention that some tools don't allow you to mix domainTypes that way, too. I'm agnostic about pseudo-primitives. I think that the "ordinal" concept suffices and is simpler. I had proposed pseudo-primitives mainly as a for formal alternative to "ordinal" order dependencies for specifying mutually exclusive domains. But VCell doesn't currently use pseudo-primitives. Through the "ordinal" ordering and use of "relative complements" you could always "subtract" the objects with prior ordinals to remove object overlap. Well, since nobody else has implemented pseudo-primitive support either, it sounds like we should just get rid of it, especially if (as you say) the same idea can be accomplished by other means. I think that a coordinate system should be time independent. I also think that object shapes/locations should be time independent also for now. Motion should probably be introduced in a subsequent kinematics extension (things get a lot more complicated) and remember that the geometries are not meshes, they are descriptive. Motions are most generally specified by velocity fields. If moving objects are strongly desired by some groups, we can discuss. Dynamic spatial is probably a reasonable next step, but I think it's safely beyond the scope of this proposal. I didn't understand what you put in the spreadsheet for this: "Yes - with boundary velocity introduced'? Can you explain? If we are going to compress the geometric representation and stick it in XML, wouldn't base64 make sense? What is the alternative? Uncompressed text can get pretty large for a sampled field or parametric geometry. It would make sense, it's just not what anyone's implemented support for yet. People have indeed implemented support for other compression, though ('deflation'). You didn't say whether or not VCell implemented quadrilateral mesh support--do you? If not, we can simplify to just triangles, since nobody else has support for that, either. I'm working through some of the example models now (and am tripping over some of VCell's limitations ... e.g. a single very small "vesicle" in a large span of x,y,z is missed by VCell's topology algorithm ... working on it). Thanks, that's perfect! -Lucian ------------------------------------------------------------------------------ _______________________________________________ sbml-spatial mailing list sbm...@li...<mailto:sbm...@li...> https://lists.sourceforge.net/lists/listinfo/sbml-spatial |