Release of plugins
1 view
Skip to first unread message
Gábor Bakos
Aug 21, 2009, 7:06:03 AM8/21/09
to HiTS users
Hello,
I am happy to tell you that yesterday there was a release from the contained plugins. You can check the details in the Changelog, although I copy the relevant parts here too (at the end of the message).
This was a long delayed release, sorry for the caused inconveniences. The good news that it should be OK, tested a bit more than usually. The new interoperability feature is not really matured, so any feedback is welcome.
I plan to add more workflows to the samples, if you need more, just tell me what would you like.
Bests, gabor
PS.: For those who are using eclipse 3.4/3.5 based KNIME a bit bad news that the helper features of the update site has been deleted, so now only the compatibility version is available. I am not sure it will reappear in the near future. (Sorry, it caused too much trouble and the release took approx. 10 hours without that trouble too, so I decided to no longer support.)
Never interrupt your enemy when he is making a mistake. ~~~ Napoleon
I am happy to tell you that yesterday there was a release from the contained plugins. You can check the details in the Changelog, although I copy the relevant parts here too (at the end of the message).
This was a long delayed release, sorry for the caused inconveniences. The good news that it should be OK, tested a bit more than usually. The new interoperability feature is not really matured, so any feedback is welcome.
I plan to add more workflows to the samples, if you need more, just tell me what would you like.
Bests, gabor
PS.: For those who are using eclipse 3.4/3.5 based KNIME a bit bad news that the helper features of the update site has been deleted, so now only the compatibility version is available. I am not sure it will reappear in the near future. (Sorry, it caused too much trouble and the release took approx. 10 hours without that trouble too, so I decided to no longer support.)
HiTS
Changes from 0.3.0 to 0.4.0
- Scores to p-values allows to compute P(Z<=x) (the Erf function) with an estamated frequency from a population.
- Issue 11 - configuration is now available in cellHTS2's outport
- Issue 22 - variance adjustment works again
- Issue 23 - contents no longer mixed in results
- Issue 25 - more detailed error messages
New features:
- The export of colour legend is now working better, more options present.
- It is possible to suppress the generation of HTML report in cellHTS2.
- Support for 384 well plates in Plate Heatmap, and the other nodes.
- Small bug fixes.
New nodes in exp (0.4.0) plugin:
- Leaf Ordering - computes the optimal leaf ordering for a hierarchical clustered tree.
- Reverse Order - changes the ordering of the tree leaves to the opposite.
KNIME utilities
Changes from 0.1.0 to 0.2.0
Created icons for those that did not have.
- Subsets - generates subsets of the selected columns.
Interoperability
Version 0.0.1
- BioConverter? - converts between different column formats
- Plate Format - converts between different plate layouts.
Never interrupt your enemy when he is making a mistake. ~~~ Napoleon
gbonamy
Aug 31, 2009, 7:49:35 PM8/31/09
to HiTS users
Gabor,
I wanted to let you know that I did the install from the update site
on Eclipse 3.5 and everything worked well. No upate with compatibility
mode were requested.
I like these new features and Icon. Great job on the smooth
transition.
2 comments:
1) Should the BioCOnverter and PlateFormat Nodes be placed in the HiTS
folder?
2) Should Pivot/Unpivot/Subsets be in DataManipulation\Column and
DirectProduct be in DataManipulation\Matrix?
Best,
Ghislain
I wanted to let you know that I did the install from the update site
on Eclipse 3.5 and everything worked well. No upate with compatibility
mode were requested.
I like these new features and Icon. Great job on the smooth
transition.
2 comments:
1) Should the BioCOnverter and PlateFormat Nodes be placed in the HiTS
folder?
2) Should Pivot/Unpivot/Subsets be in DataManipulation\Column and
DirectProduct be in DataManipulation\Matrix?
Best,
Ghislain
aborg
Sep 1, 2009, 5:20:07 AM9/1/09
to HiTS users
Hi Ghislain,
Thanks for your comments. Great to read that you had no problems.
I will put the BioConverter and Plate Format nodes to the HiTS folder.
(I initially thought of putting those to the HCDC's HTS/HCS Library
handling folder, but I guess you are right about them.)
I was not sure about the Unpivot, because I know that the developers
of KNIME also creating a new node for unpivoting (expected in KNIME
2.1), and I thought it would be confusing if there were Pivot,
Pivoting, Unpivot, Unpivoting (?) nodes in one place. (And actually I
thought that Pivot, Unpivot are more like matrix operations.) Maybe
those should go under HiTS folder too?
I think you are right about Subsets, and Direct Product, I will move
them according to your suggestion.
Thanks, gabor
Thanks for your comments. Great to read that you had no problems.
I will put the BioConverter and Plate Format nodes to the HiTS folder.
(I initially thought of putting those to the HCDC's HTS/HCS Library
handling folder, but I guess you are right about them.)
I was not sure about the Unpivot, because I know that the developers
of KNIME also creating a new node for unpivoting (expected in KNIME
2.1), and I thought it would be confusing if there were Pivot,
Pivoting, Unpivot, Unpivoting (?) nodes in one place. (And actually I
thought that Pivot, Unpivot are more like matrix operations.) Maybe
those should go under HiTS folder too?
I think you are right about Subsets, and Direct Product, I will move
them according to your suggestion.
Thanks, gabor
gbonamy
Sep 1, 2009, 2:15:17 PM9/1/09
to HiTS users
I agree that Pivot/Unpivot bellong to the matrix operation. However, I
believe these two features match the One2Many and Many2One features
that already exist in DataManipulation\Column (maybe not the best
place), which is why I made this suggestion.
I suppose that when HITS and HCDC come up with a solution, the nodes
can be merged with each other in more relevant locations. At this
point, KNIME node organization is not necessarily finalized anyways.
Best,
Ghislain
believe these two features match the One2Many and Many2One features
that already exist in DataManipulation\Column (maybe not the best
place), which is why I made this suggestion.
I suppose that when HITS and HCDC come up with a solution, the nodes
can be merged with each other in more relevant locations. At this
point, KNIME node organization is not necessarily finalized anyways.
Best,
Ghislain
Reply all
Reply to author
Forward
0 new messages