gsoc teleconference 31st May 2011

Jalview Developers
1

Hi all.

Here are the points covered in our GSOC project teleconference today:

* Varna source
  -> Work on 3.8b version : contact Yann to check on whether he can be on hand to fix up bugs/code quickly
  -> aim to include 3.8(b) with the version of Jalview that integrates varna.
  ==> Either wait on Yann to inform us of a release date - or include a copy of the beta's source with Jalview's source distribution until v3.8 of VARNA is available on the web.
  -> developing/identifying varna API : Jan + Yann - invite Yann to Jalview-dev.
  -> follow example shown on demo page - interactive update of secondary structure diagram when contacts are edited, and interactive feedback if secondary structure brackets do not match up correctly

* Data model - probably ok for the moment - Jan examined the VARNA datamodel (which uses a list of pairs indexed by sequence position - so less space efficient, but more time efficient), and also checked out if there were any scaling issues (with number of sequences) with jalview's existing one, and didn't find anything serious. However, he's prepared to have to modify the strategy in the future.

* Annotation undo
  - Jan's put in a new bug regarding undoing annotation edits
  - Jim said Jan will also need to create a new jalview.commands.<annotation edit> set of classes that allows jalview to do/undo an annotation edit after it is made.

* Annotation IO - jalview.io.AnnotationFile
  - Jan: bug hunt/test case: He needs to ensure that the RNA secondary structure parsing routines are called when annotation is imported: ie. export ss annotation, import on to new bare alignment, ensure that the new annotation is processed as RNA pair annotation corectly.

Next teleconference
Jan is unavailable at the normal time next week, so we will postpone our weekly meeting until the following week. However, he should still make his report to the NESCEnt list. We'll communicate via email for the intervening time.

Thanks for joining - it was a good talk, despite the connectivity problems :slight_smile:
Jim.

2

Hi all,

Today's GSOC project phone conference summary:

Jan made some progress with bug fixing, the changes are committed into the repository. He has updated all the relevant bugs in the bug tracker too.

He did not have much time to work on GSOC projects last week, he hoped to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the last week to NESCent.

This week Ian will be shifting his main focus from the bug fixing to VARNA integration.

Next teleconference will be at
1100am BST, 20th of July.
We'll communicate via email for the intervening time.

Regards,
Peter

Hi all.

Here are the points covered in our GSOC project

teleconference

today:

* Varna source -> Work on 3.8b version : contact Yann to

check on

whether he can be on hand to fix up bugs/code quickly ->

aim to

include 3.8(b) with the version of Jalview that integrates

varna. ==>

Either wait on Yann to inform us of a release date - or

include a

copy of the beta's source with Jalview's source distribution

until

v3.8 of VARNA is available on the web. ->

developing/identifying

varna API : Jan + Yann - invite Yann to Jalview-dev. ->

follow

example shown on demo page - interactive update of secondary

structure diagram when contacts are edited, and interactive

feedback

if secondary structure brackets do not match up correctly

* Data model - probably ok for the moment - Jan examined the

VARNA

datamodel (which uses a list of pairs indexed by sequence

position -

so less space efficient, but more time efficient), and also

checked

out if there were any scaling issues (with number of

sequences) with

jalview's existing one, and didn't find anything serious.

However,

he's prepared to have to modify the strategy in the future.

* Annotation undo - Jan's put in a new bug regarding undoing

annotation edits - Jim said Jan will also need to create a

new

jalview.commands.<annotation edit> set of classes that

allows jalview

to do/undo an annotation edit after it is made.

* Annotation IO - jalview.io.AnnotationFile - Jan: bug

hunt/test

case: He needs to ensure that the RNA secondary structure

parsing

routines are called when annotation is imported: ie. export

ss

annotation, import on to new bare alignment, ensure that the

new

annotation is processed as RNA pair annotation corectly.

Next teleconference Jan is unavailable at the normal time

next week,

so we will postpone our weekly meeting until the following

week.

However, he should still make his report to the NESCEnt list.

We'll

communicate via email for the intervening time.

Thanks for joining - it was a good talk, despite the

connectivity

···

On 30/05/2011 14:00, Jim Procter wrote:

problems :slight_smile: Jim.

3

Hi,

thank you for the summary Peter.
Could you ask Jim if the time for the next call is possible for him?
Maybe he is stuck in some ash clouds :slight_smile:

Do you know if there are any news about the problems with the git
repository? I still have to do the commit from the command line.

I created a story (JAL-842) in the 'Jalview issue tracker' for
integrating a VARNA window to Jalview. If anybody has comments about it
he could leave it there.

best
Jan

···

On 06/14/11 00:46, Peter Troshin wrote:

Hi all,

Today's GSOC project phone conference summary:

Jan made some progress with bug fixing, the changes are committed into
the repository. He has updated all the relevant bugs in the bug tracker
too.

He did not have much time to work on GSOC projects last week, he hoped
to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the
last week to NESCent.

This week Ian will be shifting his main focus from the bug fixing to
VARNA integration.

Next teleconference will be at
1100am BST, 20th of July.
We'll communicate via email for the intervening time.

Regards,
Peter

>On 30/05/2011 14:00, Jim Procter wrote:

Hi all.

Here are the points covered in our GSOC project

teleconference

today:

* Varna source -> Work on 3.8b version : contact Yann to

check on

whether he can be on hand to fix up bugs/code quickly ->

aim to

include 3.8(b) with the version of Jalview that integrates

varna. ==>

Either wait on Yann to inform us of a release date - or

include a

copy of the beta's source with Jalview's source distribution

until

v3.8 of VARNA is available on the web. ->

developing/identifying

varna API : Jan + Yann - invite Yann to Jalview-dev. ->

follow

example shown on demo page - interactive update of secondary

structure diagram when contacts are edited, and interactive

feedback

if secondary structure brackets do not match up correctly

* Data model - probably ok for the moment - Jan examined the

VARNA

datamodel (which uses a list of pairs indexed by sequence

position -

so less space efficient, but more time efficient), and also

checked

out if there were any scaling issues (with number of

sequences) with

jalview's existing one, and didn't find anything serious.

However,

he's prepared to have to modify the strategy in the future.

* Annotation undo - Jan's put in a new bug regarding undoing

annotation edits - Jim said Jan will also need to create a

new

jalview.commands.<annotation edit> set of classes that

allows jalview

to do/undo an annotation edit after it is made.

* Annotation IO - jalview.io.AnnotationFile - Jan: bug

hunt/test

case: He needs to ensure that the RNA secondary structure

parsing

routines are called when annotation is imported: ie. export

ss

annotation, import on to new bare alignment, ensure that the

new

annotation is processed as RNA pair annotation corectly.

Next teleconference Jan is unavailable at the normal time

next week,

so we will postpone our weekly meeting until the following

week.

However, he should still make his report to the NESCEnt list.

We'll

communicate via email for the intervening time.

Thanks for joining - it was a good talk, despite the

connectivity

problems :slight_smile: Jim.

_______________________________________________
Jalview-dev mailing list
Jalview-dev@jalview.org
http://www.compbio.dundee.ac.uk/mailman/listinfo/jalview-dev

4

thank you for the summary Peter.

seconded.

Could you ask Jim if the time for the next call is possible for him?
Maybe he is stuck in some ash clouds :slight_smile:

I presume the next call was scheduled for 20th June (not July as in the email?) - I'm not going to be able to make this, but I suggest you go ahead with the teleconference at that time. I'd also suggest Jan and I have another teleconference on Thursday 23rd June (11am German time ok?) so he can get me up to date.

I created a story (JAL-842) in the 'Jalview issue tracker' for
integrating a VARNA window to Jalview. If anybody has comments about it
he could leave it there.

ok. Bear in mind that commits won't appear until they hit the main jalview development branch. Something that will need to be fixed in the jira+jgit set up soon.
Jan made some progress with bug fixing, the changes are committed into

He did not have much time to work on GSOC projects last week, he hoped
to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the
last week to NESCent.

Ok. Please keep up communications on the mailing list, too ! It's particularly difficult to spot any potential problems with code when we can't see what's going on :wink:

This week Ian will be shifting his main focus from the bug fixing to
VARNA integration.

Jan - could you fill in some more detail in your phylosoc project plan about this ? What classes need to be created, etc. Send an email to the dev list when you've done that.

Jim.

···

On 14/06/2011 09:17, Jan Engelhardt wrote:

5

Hi Jim,

I hope you enjoy your holiday.
I got a code question for you, just if it's easy enough to answer per mail.

I think I roughly understand how Jmol is called via AppJmol and
AppJmolBinding. I just don't really see where the initial call to start
this cascade is. In the Desktop-class is a function called getJmols()
but according to the eclipse-reference-function it is not used anywhere.
Is it used for anything and if yes how is it called?
Where do I have to put the initial call for Varna/Jmol?

I'm able to call VARNA from the Jalview project now, so as far as I
understand the main task is to make it compatible to the
jalview.gui.Desktop.addInternalFrame function.

best
Jan

6

thank you for the summary Peter.

seconded.

Could you ask Jim if the time for the next call is possible for him?
Maybe he is stuck in some ash clouds :slight_smile:

I presume the next call was scheduled for 20th June (not July as in the
email?) - I'm not going to be able to make this, but I suggest you go
ahead with the teleconference at that time. I'd also suggest Jan and I
have another teleconference on Thursday 23rd June (11am German time ok?)
so he can get me up to date.

That should be fine.

I created a story (JAL-842) in the 'Jalview issue tracker' for
integrating a VARNA window to Jalview. If anybody has comments about it
he could leave it there.

ok. Bear in mind that commits won't appear until they hit the main
jalview development branch. Something that will need to be fixed in the
jira+jgit set up soon.

I already see both of my commits in the git history of this story...if
thats what you mean.

Jan made some progress with bug fixing, the changes are committed into

He did not have much time to work on GSOC projects last week, he hoped
to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the
last week to NESCent.

Ok. Please keep up communications on the mailing list, too ! It's
particularly difficult to spot any potential problems with code when we
can't see what's going on :wink:

I just did.

This week Ian will be shifting his main focus from the bug fixing to
VARNA integration.

Jan - could you fill in some more detail in your phylosoc project plan
about this ? What classes need to be created, etc. Send an email to the
dev list when you've done that.

ok

Jan

···

On 06/16/11 00:52, Jim Procter wrote:

On 14/06/2011 09:17, Jan Engelhardt wrote:

7

Hi Jan.

I hope you enjoy your holiday.

it's not too bad - just about to get on a train that traverses half of New Zealand's north island... and the weather has been great!

I got a code question for you, just if it's easy enough to answer per mail.

no worries. email is better for this, quite often.

I think I roughly understand how Jmol is called via AppJmol and
AppJmolBinding. I just don't really see where the initial call to start
this cascade is. In the Desktop-class is a function called getJmols()
but according to the eclipse-reference-function it is not used anywhere.
Is it used for anything and if yes how is it called?

trust the 'find references' function in this case. The only time eclipse gets it wrong is when a method is called via java's reflect interface. if you look at the code, you'll see all that Desktop.getJmols() does is returns all child windows that are instances of jalview.gui.AppJmol - which are the parent frames of any Jmol displays shown in the Jalview desktop.

Where do I have to put the initial call for Varna/Jmol?

All this happens in the constructor for jalview.gui.AppJmol - which is, of course, invoked when the user wants to open a structure view - when they select the 'view structure' entry for a PDB structure from the jalview.gui.PopupMenu instance opened for a sequence in an alignment window.

Typically, the jalview.*gui.App*Jmol class is initialised with a SequenceI, a corresponding vector PdbEntry, and one or more align panels (which could actually be null). If you follow the execution flow in the constructor, you'll see that there are basically two ways it goes:
1. there is already an instance of Jmol open, and it is showing one or all of the structures that were passed through via the PDBEntry vector (the primary or unique key for a PDBEntry's pdb structure is its PDBEntry.getFile() filename). If so - the user is prompted with a dialog box enquiring if they want to add structures or link the given alignmentPanel to an existing Jmol view - if they say no, then step 2 is invoked.
2. no jmols correspond to any of the PDBEntry's passed to the constructor, or the user wanted to open a new Jmol window. In this case, a new thread is created to load any PDB files needed and then instantiate the Jmol viewer (check out the AppJmol.paint() method - it's got a conditional that shows either a 'loading' message or eventually delegates to a JmolViewer's own paint() method).

I'm able to call VARNA from the Jalview project now, so as far as I
understand the main task is to make it compatible to the
jalview.gui.Desktop.addInternalFrame function.

Kind of, but not really - jalview.gui.Desktop is a window manager for all alignment views + sub windows - addInternalFrame adds the necessary handlers to allow the InternalFrame to be properly managed by the desktop window, and to ensure it receives events from the outside world (ie drags, and that it gets closed when the application shuts down).

What you need to do is storyboard how a varna window gets opened by the user, and what kind of events/capabilities it should handle. You can then create a class in jalview.gui that implements this, along with any special menu items needed (colourschemes, etc being tranferred from the jalview alignment window), that handles the creation and updating of a varna panel displayed within a JInternalFrame, that will close when the desktop closes, and be ready to handle/emit mouseovers generated or received from the jalview.structure.StructureManager event router.

I'm going to be out of email contact for the next 2-3 days. Assuming I get to a network connection, then I'll try to reply to any emails on Monday evening, after your teleconference with Peter. Otherwise, keep the dialog up and I'll definitely send you something on Wednesday afternoon/evening.

Jim.

···

On 16/06/2011 08:59, Jan Engelhardt wrote:

8

Hi Peter,

do you think it would be possible to have the telephone conference an
hour later, at 12am BST, 20th of July.

regards
Jan

···

On 06/14/11 00:46, Peter Troshin wrote:

Hi all,

Today's GSOC project phone conference summary:

Jan made some progress with bug fixing, the changes are committed into
the repository. He has updated all the relevant bugs in the bug tracker
too.

He did not have much time to work on GSOC projects last week, he hoped
to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the
last week to NESCent.

This week Ian will be shifting his main focus from the bug fixing to
VARNA integration.

Next teleconference will be at
1100am BST, 20th of July.
We'll communicate via email for the intervening time.

Regards,
Peter

>On 30/05/2011 14:00, Jim Procter wrote:

Hi all.

Here are the points covered in our GSOC project

teleconference

today:

* Varna source -> Work on 3.8b version : contact Yann to

check on

whether he can be on hand to fix up bugs/code quickly ->

aim to

include 3.8(b) with the version of Jalview that integrates

varna. ==>

Either wait on Yann to inform us of a release date - or

include a

copy of the beta's source with Jalview's source distribution

until

v3.8 of VARNA is available on the web. ->

developing/identifying

varna API : Jan + Yann - invite Yann to Jalview-dev. ->

follow

example shown on demo page - interactive update of secondary

structure diagram when contacts are edited, and interactive

feedback

if secondary structure brackets do not match up correctly

* Data model - probably ok for the moment - Jan examined the

VARNA

datamodel (which uses a list of pairs indexed by sequence

position -

so less space efficient, but more time efficient), and also

checked

out if there were any scaling issues (with number of

sequences) with

jalview's existing one, and didn't find anything serious.

However,

he's prepared to have to modify the strategy in the future.

* Annotation undo - Jan's put in a new bug regarding undoing

annotation edits - Jim said Jan will also need to create a

new

jalview.commands.<annotation edit> set of classes that

allows jalview

to do/undo an annotation edit after it is made.

* Annotation IO - jalview.io.AnnotationFile - Jan: bug

hunt/test

case: He needs to ensure that the RNA secondary structure

parsing

routines are called when annotation is imported: ie. export

ss

annotation, import on to new bare alignment, ensure that the

new

annotation is processed as RNA pair annotation corectly.

Next teleconference Jan is unavailable at the normal time

next week,

so we will postpone our weekly meeting until the following

week.

However, he should still make his report to the NESCEnt list.

We'll

communicate via email for the intervening time.

Thanks for joining - it was a good talk, despite the

connectivity

problems :slight_smile: Jim.

_______________________________________________
Jalview-dev mailing list
Jalview-dev@jalview.org
http://www.compbio.dundee.ac.uk/mailman/listinfo/jalview-dev

9

Hi Ian,

No problems, talk to you at 12am today.

Peter

···

On 19/06/2011 16:08, Jan Engelhardt wrote:

Hi Peter,

do you think it would be possible to have the telephone conference an
hour later, at 12am BST, 20th of July.

regards
Jan

On 06/14/11 00:46, Peter Troshin wrote:

Hi all,

Today's GSOC project phone conference summary:

Jan made some progress with bug fixing, the changes are committed into
the repository. He has updated all the relevant bugs in the bug tracker
too.

He did not have much time to work on GSOC projects last week, he hoped
to catch up with the plan this week.
Ian has written up and submitted the detailed progress report for the
last week to NESCent.

This week Ian will be shifting his main focus from the bug fixing to
VARNA integration.

Next teleconference will be at
1100am BST, 20th of July.
We'll communicate via email for the intervening time.

Regards,
Peter

  >On 30/05/2011 14:00, Jim Procter wrote:

  Hi all.
  Here are the points covered in our GSOC project

teleconference

  today:
  * Varna source -> Work on 3.8b version : contact Yann to

check on

  whether he can be on hand to fix up bugs/code quickly ->

aim to

  include 3.8(b) with the version of Jalview that integrates

varna. ==>

  Either wait on Yann to inform us of a release date - or

include a

  copy of the beta's source with Jalview's source distribution

until

  v3.8 of VARNA is available on the web. ->

developing/identifying

  varna API : Jan + Yann - invite Yann to Jalview-dev. ->

follow

  example shown on demo page - interactive update of secondary

  structure diagram when contacts are edited, and interactive

feedback

  if secondary structure brackets do not match up correctly
  * Data model - probably ok for the moment - Jan examined the

VARNA

  datamodel (which uses a list of pairs indexed by sequence

position -

  so less space efficient, but more time efficient), and also

checked

  out if there were any scaling issues (with number of

sequences) with

  jalview's existing one, and didn't find anything serious.

However,

  he's prepared to have to modify the strategy in the future.
  * Annotation undo - Jan's put in a new bug regarding undoing
  annotation edits - Jim said Jan will also need to create a

new

  jalview.commands.<annotation edit> set of classes that

allows jalview

  to do/undo an annotation edit after it is made.
  * Annotation IO - jalview.io.AnnotationFile - Jan: bug

hunt/test

  case: He needs to ensure that the RNA secondary structure

parsing

  routines are called when annotation is imported: ie. export

ss

  annotation, import on to new bare alignment, ensure that the

new

  annotation is processed as RNA pair annotation corectly.
  Next teleconference Jan is unavailable at the normal time

next week,

  so we will postpone our weekly meeting until the following

week.

  However, he should still make his report to the NESCEnt list.

We'll

  communicate via email for the intervening time.
  Thanks for joining - it was a good talk, despite the

connectivity

  problems :slight_smile: Jim.

_______________________________________________
Jalview-dev mailing list
Jalview-dev@jalview.org
http://www.compbio.dundee.ac.uk/mailman/listinfo/jalview-dev

_______________________________________________
Jalview-dev mailing list
Jalview-dev@jalview.org
http://www.compbio.dundee.ac.uk/mailman/listinfo/jalview-dev