From jburdick at keyfitz.org  Tue May  8 15:15:46 2007
From: jburdick at keyfitz.org (Josh Burdick)
Date: Tue, 08 May 2007 15:15:46 -0400
Subject: [Biojava-dev] reading a subsequence from a .nib file
In-Reply-To: <OF5267BEF4.5CB10CC3-ON482572C3.000EEC6E-482572C3.000F17B0@ah.novartis.com>
References: <OF5267BEF4.5CB10CC3-ON482572C3.000EEC6E-482572C3.000F17B0@ah.novartis.com>
Message-ID: <1178651747.1738.137.camel@localhost.localdomain>

On Fri, 2007-04-20 at 10:44 +0800, mark.schreiber at novartis.com wrote:
> Hi Josh -
> 
> Looks good.  Just one thing, your JUnit test contains a hardcoded file 
> path to the test file which means it is not portable. Could you modify 
> that so that it loads the file from the classpath as a resource (see some 
> of the IO unit tests for examples). Can you also provide the test file.
> 
  I'm not currently compiling all of BioJava, and didn't know how to add
the file as a resource.  So I'm afraid I haven't set the pathname to
something sensible.

  But I created a small test file which is a fragment of chromosome 18.

  The relevant files are:

http://keyfitz.org/jburdick/read_nib_file_java/NibFile.java
http://keyfitz.org/jburdick/read_nib_file_java/NibFileTest.java
http://keyfitz.org/jburdick/read_nib_file_java/chr18_chunk.nib
http://keyfitz.org/jburdick/read_nib_file_java/readme.txt

  Josh

> Best regards,
> 
> - Mark
> 
> Mark Schreiber
> Research Investigator (Bioinformatics)
> 
> Novartis Institute for Tropical Diseases (NITD)
> 10 Biopolis Road
> #05-01 Chromos
> Singapore 138670
> www.nitd.novartis.com
> 
> phone +65 6722 2973
> fax  +65 6722 2910
> 

[...]



From stephen at blackrim.net  Wed May  9 11:45:34 2007
From: stephen at blackrim.net (Stephen A Smith)
Date: Wed, 09 May 2007 11:45:34 -0400
Subject: [Biojava-dev] [Off Topic] Google group
Message-ID: <4641EC9E.4010307@blackrim.net>

Hi all,
Just letting you know there is a google group open now for discussions 
of all thing programming and evolutionary biology. You can find it here 
http://groups.google.com/group/evo_code.
Figured the people at bio* might be interested.
Take care
Stephen Smith

From darin.london at duke.edu  Thu May 10 12:17:38 2007
From: darin.london at duke.edu (darin.london at duke.edu)
Date: Thu, 10 May 2007 12:17:38 -0400
Subject: [Biojava-dev] BOSC 2007 Second Call For Papers
Message-ID: <200705101617.l4AGHc7L002450@tenero.duhs.duke.edu>


The BOSC Organizing Committee are proud to announce BOSC 2007, occurring
in Vienna, Austria on July 19th, 20th.  The conference this year
promises to be exciting, as the BOSC developers attempt to define and
solve currently intractable problems in Bioinformatics. Please refer to
the following website for complete information, and requests for
submissions.   Thank you, and we hope to see you in Vienna.

http://open-bio.org/wiki/BOSC_2007


The BOSC organizing Committee


Please pass this email on to anyone that would be interested.



			 


From jeff at dvss.net  Thu May 10 21:30:11 2007
From: jeff at dvss.net (Jeff Szielenski)
Date: Thu, 10 May 2007 17:30:11 -0800
Subject: [Biojava-dev] Project Contribution.
Message-ID: <003901c7936b$ecd256e0$0601a8c0@laptop>

Hello,
 
I am currently studying bioinformatics at UIC. I am interested in
joining an open source project to start applying my knowledge. What if
any help do you guys need? I have a software engineering background and
just finished an introductory course in bioinformatics.
 
Jeff
 

From markjschreiber at gmail.com  Thu May 10 22:54:36 2007
From: markjschreiber at gmail.com (Mark Schreiber)
Date: Fri, 11 May 2007 10:54:36 +0800
Subject: [Biojava-dev] Project Contribution.
In-Reply-To: <003901c7936b$ecd256e0$0601a8c0@laptop>
References: <003901c7936b$ecd256e0$0601a8c0@laptop>
Message-ID: <93b45ca50705101954j3f899f49r438fe8d587ba027@mail.gmail.com>

Hi -

We need volunteers to convert the example programs in the biojava.org
cookbook to use the newer biojavax API's.

- Mark

On 5/11/07, Jeff Szielenski <jeff at dvss.net> wrote:
> Hello,
>
> I am currently studying bioinformatics at UIC. I am interested in
> joining an open source project to start applying my knowledge. What if
> any help do you guys need? I have a software engineering background and
> just finished an introductory course in bioinformatics.
>
> Jeff
>
> _______________________________________________
> biojava-dev mailing list
> biojava-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-dev
>

From markjschreiber at gmail.com  Sun May 13 23:14:43 2007
From: markjschreiber at gmail.com (Mark Schreiber)
Date: Mon, 14 May 2007 11:14:43 +0800
Subject: [Biojava-dev] DOM, JDom, Xerces?
Message-ID: <93b45ca50705132014n4b0ed35eu67153def42ac705c@mail.gmail.com>

Hello -

I am looking into XPath and XQuery as a way rapidly process the
increasing number of bioinformatics XML formats. As is typical with
Java there is a too many APIs to choose from.

Have people had good experiences with the internal java API (if so
JDK1.5 or 1.6), or is JDom better (even though it is not really DOM)?
How about Xerces and Xalan from Apache???

Is there much support for XQuery yet? I have seen there is a JSR for
official support of XQuery, is there an associated API for download?

Any comments would be greatly appreciated.

- Mark

From ayates at ebi.ac.uk  Mon May 14 04:47:28 2007
From: ayates at ebi.ac.uk (Andy Yates)
Date: Mon, 14 May 2007 09:47:28 +0100
Subject: [Biojava-dev] DOM, JDom, Xerces?
In-Reply-To: <93b45ca50705132014n4b0ed35eu67153def42ac705c@mail.gmail.com>
References: <93b45ca50705132014n4b0ed35eu67153def42ac705c@mail.gmail.com>
Message-ID: <38571708-E02B-4B75-A3AC-3E0E090DBB77@ebi.ac.uk>

Hey Mark,

I've tried using XPath before on a couple of projects & it is quite  
nice. In terms of implementations the internal Java one isn't that  
bad but I really like XOM. It's a very clear/clean API and is quite  
easy to switch between using XPath back into the normal API.

The other program I've heard reasonable things about is http:// 
javadude.com/tools/antxr/index.html which is the XML equivalent to  
antlr. I haven't used it yet but it does look interesting

Andy

On 14 May 2007, at 04:14, Mark Schreiber wrote:

> Hello -
>
> I am looking into XPath and XQuery as a way rapidly process the
> increasing number of bioinformatics XML formats. As is typical with
> Java there is a too many APIs to choose from.
>
> Have people had good experiences with the internal java API (if so
> JDK1.5 or 1.6), or is JDom better (even though it is not really DOM)?
> How about Xerces and Xalan from Apache???
>
> Is there much support for XQuery yet? I have seen there is a JSR for
> official support of XQuery, is there an associated API for download?
>
> Any comments would be greatly appreciated.
>
> - Mark
> _______________________________________________
> biojava-dev mailing list
> biojava-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-dev


From darin.london at duke.edu  Mon May 14 10:44:56 2007
From: darin.london at duke.edu (darin.london at duke.edu)
Date: Mon, 14 May 2007 10:44:56 -0400
Subject: [Biojava-dev] BOSC 2007 Abstract Submission Deadline Extended
Message-ID: <200705141444.l4EEiuxQ026957@tenero.duhs.duke.edu>


Due to technical difficulties in sending out the 2nd call for papers, the BOSC organizers are extending the deadline for abstract submissions to Monday May 21st.  The announcement day will remain the same so that it remains before the Early Discount Date.

http://open-bio.org/wiki/BOSC_2007


The BOSC organizing Committee


Please pass this email on to anyone that would be interested.



			 


From phidias51 at gmail.com  Mon May 14 12:00:07 2007
From: phidias51 at gmail.com (Mark Fortner)
Date: Mon, 14 May 2007 09:00:07 -0700
Subject: [Biojava-dev] DOM, JDom, Xerces?
In-Reply-To: <38571708-E02B-4B75-A3AC-3E0E090DBB77@ebi.ac.uk>
References: <93b45ca50705132014n4b0ed35eu67153def42ac705c@mail.gmail.com>
	<38571708-E02B-4B75-A3AC-3E0E090DBB77@ebi.ac.uk>
Message-ID: <6e1d61f50705140900w1df2d5d7l5f39fb418668f15f@mail.gmail.com>

Mark & Andy,
I haven't used XOM before, but it sounds interesting.  I HAVE used Saxon and
it has good support for both XPath and XQuery.

Regards,

Mark Fortner

On 5/14/07, Andy Yates <ayates at ebi.ac.uk> wrote:
>
> Hey Mark,
>
> I've tried using XPath before on a couple of projects & it is quite
> nice. In terms of implementations the internal Java one isn't that
> bad but I really like XOM. It's a very clear/clean API and is quite
> easy to switch between using XPath back into the normal API.
>
> The other program I've heard reasonable things about is http://
> javadude.com/tools/antxr/index.html which is the XML equivalent to
> antlr. I haven't used it yet but it does look interesting
>
> Andy
>
> On 14 May 2007, at 04:14, Mark Schreiber wrote:
>
> > Hello -
> >
> > I am looking into XPath and XQuery as a way rapidly process the
> > increasing number of bioinformatics XML formats. As is typical with
> > Java there is a too many APIs to choose from.
> >
> > Have people had good experiences with the internal java API (if so
> > JDK1.5 or 1.6), or is JDom better (even though it is not really DOM)?
> > How about Xerces and Xalan from Apache???
> >
> > Is there much support for XQuery yet? I have seen there is a JSR for
> > official support of XQuery, is there an associated API for download?
> >
> > Any comments would be greatly appreciated.
> >
> > - Mark
> > _______________________________________________
> > biojava-dev mailing list
> > biojava-dev at lists.open-bio.org
> > http://lists.open-bio.org/mailman/listinfo/biojava-dev
>
> _______________________________________________
> biojava-dev mailing list
> biojava-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-dev
>

From ayates at ebi.ac.uk  Mon May 14 12:11:16 2007
From: ayates at ebi.ac.uk (Andy Yates)
Date: Mon, 14 May 2007 17:11:16 +0100
Subject: [Biojava-dev] DOM, JDom, Xerces?
In-Reply-To: <6e1d61f50705140900w1df2d5d7l5f39fb418668f15f@mail.gmail.com>
References: <93b45ca50705132014n4b0ed35eu67153def42ac705c@mail.gmail.com>
	<38571708-E02B-4B75-A3AC-3E0E090DBB77@ebi.ac.uk>
	<6e1d61f50705140900w1df2d5d7l5f39fb418668f15f@mail.gmail.com>
Message-ID: <FF0B482A-103D-4ED3-ADAD-23BDBB7E1612@ebi.ac.uk>

Hi Mark,

I've not used Saxon before but from what I can take from the docs it  
seems quite nice. However if anyone is going to do some XML  
processing give it a go. Recently I used XOM's document builder and I  
have to say that I wrote the majority without a need to consult XOM's  
documentation (the only bit I needed was figuring out how to print a  
pretty XML file out). It's very powerful & very easy to read in code.

Also if you fancy having a look at some newer technologies Groovy has  
some very nice ways of parsing and generating XML. Documentation is a  
bit sparse in places but once you're in the groovy mindset (or bought  
the book) it's fine (and to be honest the Groovy builder objects are  
worth the pain). I'm moving most of my scripting to it now; it's like  
programming in Perl but with the comfort of knowing that you're in  
Java and can jump out into Java classes whenever you want ... like  
BioJava ;)

Andy

On 14 May 2007, at 17:00, Mark Fortner wrote:

> Mark & Andy,
> I haven't used XOM before, but it sounds interesting.  I HAVE used  
> Saxon and
> it has good support for both XPath and XQuery.
>
> Regards,
>
> Mark Fortner
>
> On 5/14/07, Andy Yates <ayates at ebi.ac.uk> wrote:
>>
>> Hey Mark,
>>
>> I've tried using XPath before on a couple of projects & it is quite
>> nice. In terms of implementations the internal Java one isn't that
>> bad but I really like XOM. It's a very clear/clean API and is quite
>> easy to switch between using XPath back into the normal API.
>>
>> The other program I've heard reasonable things about is http://
>> javadude.com/tools/antxr/index.html which is the XML equivalent to
>> antlr. I haven't used it yet but it does look interesting
>>
>> Andy
>>
>> On 14 May 2007, at 04:14, Mark Schreiber wrote:
>>
>>> Hello -
>>>
>>> I am looking into XPath and XQuery as a way rapidly process the
>>> increasing number of bioinformatics XML formats. As is typical with
>>> Java there is a too many APIs to choose from.
>>>
>>> Have people had good experiences with the internal java API (if so
>>> JDK1.5 or 1.6), or is JDom better (even though it is not really  
>>> DOM)?
>>> How about Xerces and Xalan from Apache???
>>>
>>> Is there much support for XQuery yet? I have seen there is a JSR for
>>> official support of XQuery, is there an associated API for download?
>>>
>>> Any comments would be greatly appreciated.
>>>
>>> - Mark
>>> _______________________________________________
>>> biojava-dev mailing list
>>> biojava-dev at lists.open-bio.org
>>> http://lists.open-bio.org/mailman/listinfo/biojava-dev
>>
>> _______________________________________________
>> biojava-dev mailing list
>> biojava-dev at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/biojava-dev
>>
> _______________________________________________
> biojava-dev mailing list
> biojava-dev at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-dev


From bugzilla-daemon at portal.open-bio.org  Thu May 17 04:19:45 2007
From: bugzilla-daemon at portal.open-bio.org (bugzilla-daemon at portal.open-bio.org)
Date: Thu, 17 May 2007 04:19:45 -0400
Subject: [Biojava-dev] [Bug 2297] New: XMLDistributionWriter closes
	OutputStreams
Message-ID: <bug-2297-485@http.bugzilla.open-bio.org/>

http://bugzilla.open-bio.org/show_bug.cgi?id=2297

           Summary: XMLDistributionWriter closes OutputStreams
           Product: BioJava
           Version: live (CVS source)
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: dist/dp
        AssignedTo: biojava-dev at biojava.org
        ReportedBy: mark.schreiber at novartis.com


The XMLDistributionWriter has a couple of places where it closes the output
stream after writing out a distribution. 

This is bad if you want to write more than one and really bad if you use
System.out!!


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

From bugzilla-daemon at portal.open-bio.org  Thu May 17 21:26:11 2007
From: bugzilla-daemon at portal.open-bio.org (bugzilla-daemon at portal.open-bio.org)
Date: Thu, 17 May 2007 21:26:11 -0400
Subject: [Biojava-dev] [Bug 2298] New: Use of Big Decimal in
	DistributionTrainer
Message-ID: <bug-2298-485@http.bugzilla.open-bio.org/>

http://bugzilla.open-bio.org/show_bug.cgi?id=2298

           Summary: Use of Big Decimal in DistributionTrainer
           Product: BioJava
           Version: live (CVS source)
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: enhancement
          Priority: P2
         Component: dist/dp
        AssignedTo: biojava-dev at biojava.org
        ReportedBy: mark.schreiber at novartis.com


It would be more accurate if the Distribution API used BigDecimal internally as
much as possible to minimize rounding errors. Likely places that need changing
are Distribution, DistributionTrainer, and the DP package.

An alternative would be to make a biojavax extension of these classes that has
methods that use BigDecimal directly. This would be desirable for biojava 1.6


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

From bugzilla-daemon at portal.open-bio.org  Thu May 17 23:44:41 2007
From: bugzilla-daemon at portal.open-bio.org (bugzilla-daemon at portal.open-bio.org)
Date: Thu, 17 May 2007 23:44:41 -0400
Subject: [Biojava-dev] [Bug 2297] XMLDistributionWriter closes OutputStreams
In-Reply-To: <bug-2297-485@http.bugzilla.open-bio.org/>
Message-ID: <200705180344.l4I3if2R016019@portal.open-bio.org>

http://bugzilla.open-bio.org/show_bug.cgi?id=2297


mark.schreiber at novartis.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED




------- Comment #1 from mark.schreiber at novartis.com  2007-05-17 23:44 EST -------
Fixed to no longer close streams.


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

From bugzilla-daemon at portal.open-bio.org  Fri May 18 09:54:07 2007
From: bugzilla-daemon at portal.open-bio.org (bugzilla-daemon at portal.open-bio.org)
Date: Fri, 18 May 2007 09:54:07 -0400
Subject: [Biojava-dev] [Bug 2298] Use of Big Decimal in DistributionTrainer
In-Reply-To: <bug-2298-485@http.bugzilla.open-bio.org/>
Message-ID: <200705181354.l4IDs7Xf007500@portal.open-bio.org>

http://bugzilla.open-bio.org/show_bug.cgi?id=2298


holland at ebi.ac.uk changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED




------- Comment #1 from holland at ebi.ac.uk  2007-05-18 09:54 EST -------
Do you want double changed to BigDecimal for _every_ occurrence of double (e.g.
weights, scores, etc.) or just for one of these occurrences? 

Modifying the existing classes is going to produce very inefficient code as
there is a lot of use of the public API internally, which would require
boxing/unboxing of double values to/from BigDecimal values. Such changes are
possible but would cause very slow code due to the overhead of doing all these
conversions.

Better quality code could be produced if we just rewrote the distribution code
completely with BigDecimals in mind. Extending the existing modules would not
suffice because the public API is too tightly linked with the internal
calculations, and to add in BigDecimal-awareness would be tantamount to a
complete rewrite anyway!

I think we can leave this until 1.6, but I'll leave it as open and assigned so
that we don't forget it.


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

From bugzilla-daemon at portal.open-bio.org  Thu May 24 03:10:21 2007
From: bugzilla-daemon at portal.open-bio.org (bugzilla-daemon at portal.open-bio.org)
Date: Thu, 24 May 2007 03:10:21 -0400
Subject: [Biojava-dev] [Bug 2301] New: Initialization error of
	org.biojava.bio.alignment.NeedlemanWunsch when considering
	affine gap penalty
Message-ID: <bug-2301-485@http.bugzilla.open-bio.org/>

http://bugzilla.open-bio.org/show_bug.cgi?id=2301

           Summary: Initialization error of
                    org.biojava.bio.alignment.NeedlemanWunsch when
                    considering affine gap penalty
           Product: BioJava
           Version: live (CVS source)
          Platform: PC
        OS/Version: Windows XP
            Status: NEW
          Severity: normal
          Priority: P2
         Component: alignment
        AssignedTo: biojava-dev at biojava.org
        ReportedBy: cwu2 at eecs.wsu.edu
                CC: cwu2 at eecs.wsu.edu


In the class org.biojava.bio.alignment.NeedlemanWunsch, the initilization of
the first row and first column should be the following code when considering
affine gap. Otherwise, the first CostMatrix[1][0] will not get the right value!
Thanks!
        for (i=1; i<=query.length();i++) {
//          CostMatrix[i][0]  = CostMatrix[i-1][0] + delete;
          E[i][0] = Double.POSITIVE_INFINITY;
          CostMatrix[i][0] = F[i][0] = delete + i*gapExt;
        }
        for (j=1; j<=subject.length(); j++) {
//          CostMatrix[0][j]  = CostMatrix[0][j-1] + insert;
          F[0][j] = Double.POSITIVE_INFINITY;
          CostMatrix[0][j] = E[0][j] = insert + j*gapExt;
        }


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

From rejarohit2004 at gmail.com  Tue May  1 05:14:48 2007
From: rejarohit2004 at gmail.com (rohit reja)
Date: Tue, 01 May 2007 09:14:48 -0000
Subject: [Biojava-dev] query
Message-ID: <29c042ff0705010214q4f64bbe3r5f704665cec40693@mail.gmail.com>

hello all
i m a novice programmer in java  with the knowledge of java core
i want to build up a biological database using biojava
is there need to go for J2EE before learning biojava?

Is biojava incorporated in NetBeans IDE.?

Regards
-- 
Rohit Reja
3rd -B.tech-Bioinformatics
VIT University
Vellore

From FLong at skcc.org  Tue May  8 03:23:00 2007
From: FLong at skcc.org (Fred Long)
Date: Tue, 08 May 2007 07:23:00 -0000
Subject: [Biojava-dev] fixed and streamlined SmithWaterman.java
Message-ID: <D8B5E1124CA6F7459AAD9A8C87691C0111C07E@Sidney.skcc.org>

I found that SmithWaterman was calling super() with the wrong
parameters, which caused getMatch() to return the wrong value.  While
fixing that I removed a lot of the redundancy from SmithWaterman.  Here
is the patch file is anybody is interested.  Disclaimer: I'm not a
BioJava developer.

FL

------------------------------------



*** old/NeedlemanWunsch.java	2007-05-08 00:05:20.000000000 -0700
--- NeedlemanWunsch.java	2007-05-08 00:05:38.000000000 -0700
***************
*** 63,69 ****
    protected SubstitutionMatrix subMatrix;
    protected Alignment pairalign;
    protected String alignment;
!   private double insert, delete, gapExt, match, replace;
    
    
    /** Constructs a new Object with the given parameters based on the
Needleman-Wunsch algorithm
--- 63,69 ----
    protected SubstitutionMatrix subMatrix;
    protected Alignment pairalign;
    protected String alignment;
!   protected double insert, delete, gapExt, match, replace;
    
    
    /** Constructs a new Object with the given parameters based on the
Needleman-Wunsch algorithm
***************
*** 517,523 ****
     *        target sequence
     * @return The score for the given substitution.
     */
!   private double matchReplace(Sequence query, Sequence subject, int i,
int j) {
      try {
        return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
      } catch (Exception exc) {
--- 517,523 ----
     *        target sequence
     * @return The score for the given substitution.
     */
!   protected double matchReplace(Sequence query, Sequence subject, int
i, int j) {
      try {
        return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
      } catch (Exception exc) {
*** old/SmithWaterman.java	2007-05-08 00:05:20.000000000 -0700
--- SmithWaterman.java	2007-05-08 00:05:38.000000000 -0700
***************
*** 53,141 ****
  public class SmithWaterman extends NeedlemanWunsch
  {
  
-   private double match, replace, insert, delete, gapExt;
    private double[][] scoreMatrix;  
    
  
    /** Constructs the new SmithWaterman alignment object. Alignments
are only performed,
      * if the alphabet of the given <code>SubstitutionMatrix</code>
equals the alpabet of
!     * both the query and the target <code>Sequence</code>. The
alignment parameters here
!     * are expenses and not scores as they are in the
<code>NeedlemanWunsch</code> object.
!     * scores are just given by multipliing the expenses with
<code>(-1)</code>. For example
!     * you could use parameters like "-2, 5, 3, 3, 0". If the expenses
for gap extension
      * are equal to the cost of starting a gap (delete or insert), no
affine gap penalties
      * are used, which saves memory.
      *
!     * @param match expenses for a match
!     * @param replace expenses for a replace operation
!     * @param insert expenses for a gap opening in the query sequence
!     * @param delete expenses for a gap opening in the target sequence
!     * @param gapExtend expenses for the extension of a gap which was
started earlier.
      * @param matrix the <code>SubstitutionMatrix</code> object to use.
      */
    public SmithWaterman(double match, double replace, double insert,
double delete, double gapExtend, SubstitutionMatrix matrix)
    {
!     super(insert, delete, gapExtend, match, replace, matrix);
!     this.match      = -match;
!     this.replace    = -replace;
!     this.insert     = -insert;
!     this.delete     = -delete;
!     this.gapExt     = -gapExtend;
!     this.subMatrix  = matrix;
!     this.alignment  = "";
    }
    
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for an insert operation to the specified value.
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param ins costs for a single insert operation
-     */
-   public void setInsert(double ins) {
-     this.insert = -ins;
-   }
-   
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a delete operation to the specified value. 
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param del costs for a single deletion operation
-     */
-   public void setDelete(double del) {
-     this.delete = -del;
-   }
-   
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for an extension of any gap (insert or delete)
to the 
-     * specified value.
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1. 
-     * @param ge costs for any gap extension
-     */
-   public void setGapExt(double ge) {
-     this.gapExt = -ge;
-   }
-   
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a match operation to the specified value. 
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param ma costs for a single match operation
-     */
-   public void setMatch(double ma) {
-     this.match = -ma;
-   }
-   
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a replace operation to the specified value.

-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param rep costs for a single replace operation
-     */
-   public void setReplace(double rep) {
-     this.replace = -rep;
-   }  
-   
  
    /** Overrides the method inherited from the NeedlemanWunsch and
performs only a local alignment.
      * It finds only the longest common subsequence. This is good for
the beginning, but it might
--- 53,79 ----
  public class SmithWaterman extends NeedlemanWunsch
  {
  
    private double[][] scoreMatrix;  
    
  
    /** Constructs the new SmithWaterman alignment object. Alignments
are only performed,
      * if the alphabet of the given <code>SubstitutionMatrix</code>
equals the alpabet of
!     * both the query and the target <code>Sequence</code>. If the
expenses for gap extension
      * are equal to the cost of starting a gap (delete or insert), no
affine gap penalties
      * are used, which saves memory.
      *
!     * @param match expenses for a match (usually < 0)
!     * @param replace expenses for a replace operation (usually > 0)
!     * @param insert expenses for a gap opening in the query sequence
(usually > 0)
!     * @param delete expenses for a gap opening in the target sequence
(usually > 0)
!     * @param gapExtend expenses for the extension of a gap which was
started earlier (usually >= 0)
      * @param matrix the <code>SubstitutionMatrix</code> object to use.
      */
    public SmithWaterman(double match, double replace, double insert,
double delete, double gapExtend, SubstitutionMatrix matrix)
    {
!     super(match, replace, insert, delete, gapExtend, matrix);
    }
    
  
    /** Overrides the method inherited from the NeedlemanWunsch and
performs only a local alignment.
      * It finds only the longest common subsequence. This is good for
the beginning, but it might
***************
*** 186,193 ****
          for (i=1; i<=query.length();   i++)
            for (j=1; j<=subject.length(); j++)
            {
!             E[i][j] = Math.max(E[i][j-1], scoreMatrix[i][j-1] +
insert) + gapExt;
!             F[i][j] = Math.max(F[i-1][j], scoreMatrix[i-1][j] +
delete) + gapExt;
              scoreMatrix[i][j] = max(0.0, E[i][j], F[i][j],
scoreMatrix[i-1][j-1] + matchReplace(query, subject, i, j));
              
              if (scoreMatrix[i][j] > scoreMatrix[maxI][maxJ]) {
--- 124,131 ----
          for (i=1; i<=query.length();   i++)
            for (j=1; j<=subject.length(); j++)
            {
!             E[i][j] = Math.max(E[i][j-1], scoreMatrix[i][j-1] -
insert) - gapExt;
!             F[i][j] = Math.max(F[i-1][j], scoreMatrix[i-1][j] -
delete) - gapExt;
              scoreMatrix[i][j] = max(0.0, E[i][j], F[i][j],
scoreMatrix[i-1][j-1] + matchReplace(query, subject, i, j));
              
              if (scoreMatrix[i][j] > scoreMatrix[maxI][maxJ]) {
***************
*** 223,229 ****
                // Insert || finish gap if extended gap is opened
                } else if (scoreMatrix[i][j] == E[i][j] ||
gap_extend[0]) {
                  //check if gap has been extended or freshly opened
!                 gap_extend[0] = (E[i][j] != scoreMatrix[i][j-1] +
insert + gapExt);
                  
                  align[0] = '-' + align[0];
                  align[1] = st.tokenizeSymbol(subject.symbolAt(j--)) +
align[1];
--- 161,167 ----
                // Insert || finish gap if extended gap is opened
                } else if (scoreMatrix[i][j] == E[i][j] ||
gap_extend[0]) {
                  //check if gap has been extended or freshly opened
!                 gap_extend[0] = (E[i][j] != scoreMatrix[i][j-1] -
insert - gapExt);
                  
                  align[0] = '-' + align[0];
                  align[1] = st.tokenizeSymbol(subject.symbolAt(j--)) +
align[1];
***************
*** 232,238 ****
                // Delete || finish gap if extended gap is opened
                } else {
                  //check if gap has been extended or freshly opened
!                 gap_extend[1] = (F[i][j] != scoreMatrix[i-1][j] +
delete + gapExt);
                  
                  align[0] = st.tokenizeSymbol(query.symbolAt(i--)) +
align[0];
                  align[1] = '-' + align[1];
--- 170,176 ----
                // Delete || finish gap if extended gap is opened
                } else {
                  //check if gap has been extended or freshly opened
!                 gap_extend[1] = (F[i][j] != scoreMatrix[i-1][j] -
delete - gapExt);
                  
                  align[0] = st.tokenizeSymbol(query.symbolAt(i--)) +
align[0];
                  align[1] = '-' + align[1];
***************
*** 257,264 ****
              
              scoreMatrix[i][j] = max(
                0.0,
!               scoreMatrix[i-1][j]   + delete,
!               scoreMatrix[i][j-1]   + insert,
                scoreMatrix[i-1][j-1] + matchReplace(query, subject, i,
j)
              );
                          
--- 195,202 ----
              
              scoreMatrix[i][j] = max(
                0.0,
!               scoreMatrix[i-1][j]   - delete,
!               scoreMatrix[i][j-1]   - insert,
                scoreMatrix[i-1][j-1] + matchReplace(query, subject, i,
j)
              );
                          
***************
*** 291,297 ****
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    
                  // Insert
!                 } else if (scoreMatrix[i][j] == scoreMatrix[i][j-1] +
insert) {
                    align[0] = '-' + align[0];
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    path     = ' ' + path;
--- 229,235 ----
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    
                  // Insert
!                 } else if (scoreMatrix[i][j] == scoreMatrix[i][j-1] -
insert) {
                    align[0] = '-' + align[0];
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    path     = ' ' + path;
***************
*** 382,409 ****
      if ((y > z)) return y;
      return z;
    }
-   
-   /** This method computes the scores for the substution of the i-th
symbol
-    * of query by the j-th symbol of subject. 
-    * 
-    * @param query The query sequence
-    * @param subject The target sequence
-    * @param i The position of the symbol under consideration within
the 
-    *        query sequence (starting from one)
-    * @param j The position of the symbol under consideration within
the
-    *        target sequence
-    * @return The score for the given substitution.
-    */
-   private double matchReplace(Sequence query, Sequence subject, int i,
int j) {
-     try {
-       return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
-     } catch (Exception exc) {
-       if (query.symbolAt(i).getMatches().contains(subject.symbolAt(j))
||
-         subject.symbolAt(j).getMatches().contains(query.symbolAt(i)))
-         return match;
-       return replace;
-     }
-   }
-   
-   
  }
--- 320,323 ----


From blee34 at mail.gatech.edu  Sat May 19 11:29:54 2007
From: blee34 at mail.gatech.edu (blee34 at mail.gatech.edu)
Date: Sat, 19 May 2007 11:29:54 -0400
Subject: [Biojava-dev] GSoC student
Message-ID: <1179588594.464f17f2a2f94@webmail.mail.gatech.edu>


Dear  BioJava group

Hi, My name Bohyun Lee who will be participating in BioJava project this
summer (as a part of Google Summer of Code program, primary mentor: Richard
Holland). Especially, I'm interested in Phylogeny reconstruction method and I'm
planning to build some APIs for it. (including NEXUS parser, Tree building
method, distance calculation methods, etc.)

I'll be starting my project from comming week and I just wanted to introduce
myself to the group before it starts. If any of you can take a look at my
project plan and share your thoughts about it, I would be more than happy to
listen.

I'm looking forward to be a part of BioJava group this summer and I wish a great
summer to everyone as well.

Thank you very much.

Best regards,
Bohyun Lee


-------------- next part --------------
A non-text attachment was scrubbed...
Name: BioJava_plan.doc
Type: application/msword
Size: 28672 bytes
Desc: not available
Url : http://lists.open-bio.org/pipermail/biojava-dev/attachments/20070519/2ea8ae19/attachment-0001.doc 

From mark.schreiber at novartis.com  Thu May 31 21:32:32 2007
From: mark.schreiber at novartis.com (mark.schreiber at novartis.com)
Date: Fri, 1 Jun 2007 09:32:32 +0800
Subject: [Biojava-dev] fixed and streamlined SmithWaterman.java
Message-ID: <OF85B0E51F.B6EDB513-ON482572ED.00085607-482572ED.00087910@ah.novartis.com>

Hi -

Could the person who maintains the SW code (Andreas Drager??) take a look 
at this and check it in if required?

Thanks.

- Mark





"Fred Long" <FLong at skcc.org>
Sent by: biojava-dev-bounces at lists.open-bio.org
05/08/2007 03:15 PM

 
        To:     <biojava-dev at biojava.org>
        cc:     (bcc: Mark Schreiber/GP/Novartis)
        Subject:        [Biojava-dev] fixed and streamlined SmithWaterman.java


I found that SmithWaterman was calling super() with the wrong
parameters, which caused getMatch() to return the wrong value.  While
fixing that I removed a lot of the redundancy from SmithWaterman.  Here
is the patch file is anybody is interested.  Disclaimer: I'm not a
BioJava developer.

FL

------------------------------------



*** old/NeedlemanWunsch.java             2007-05-08 00:05:20.000000000 
-0700
--- NeedlemanWunsch.java                 2007-05-08 00:05:38.000000000 
-0700
***************
*** 63,69 ****
    protected SubstitutionMatrix subMatrix;
    protected Alignment pairalign;
    protected String alignment;
!   private double insert, delete, gapExt, match, replace;
 
 
    /** Constructs a new Object with the given parameters based on the
Needleman-Wunsch algorithm
--- 63,69 ----
    protected SubstitutionMatrix subMatrix;
    protected Alignment pairalign;
    protected String alignment;
!   protected double insert, delete, gapExt, match, replace;
 
 
    /** Constructs a new Object with the given parameters based on the
Needleman-Wunsch algorithm
***************
*** 517,523 ****
     *        target sequence
     * @return The score for the given substitution.
     */
!   private double matchReplace(Sequence query, Sequence subject, int i,
int j) {
      try {
        return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
      } catch (Exception exc) {
--- 517,523 ----
     *        target sequence
     * @return The score for the given substitution.
     */
!   protected double matchReplace(Sequence query, Sequence subject, int
i, int j) {
      try {
        return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
      } catch (Exception exc) {
*** old/SmithWaterman.java               2007-05-08 00:05:20.000000000 
-0700
--- SmithWaterman.java           2007-05-08 00:05:38.000000000 -0700
***************
*** 53,141 ****
  public class SmithWaterman extends NeedlemanWunsch
  {
 
-   private double match, replace, insert, delete, gapExt;
    private double[][] scoreMatrix; 
 
 
    /** Constructs the new SmithWaterman alignment object. Alignments
are only performed,
      * if the alphabet of the given <code>SubstitutionMatrix</code>
equals the alpabet of
!     * both the query and the target <code>Sequence</code>. The
alignment parameters here
!     * are expenses and not scores as they are in the
<code>NeedlemanWunsch</code> object.
!     * scores are just given by multipliing the expenses with
<code>(-1)</code>. For example
!     * you could use parameters like "-2, 5, 3, 3, 0". If the expenses
for gap extension
      * are equal to the cost of starting a gap (delete or insert), no
affine gap penalties
      * are used, which saves memory.
      *
!     * @param match expenses for a match
!     * @param replace expenses for a replace operation
!     * @param insert expenses for a gap opening in the query sequence
!     * @param delete expenses for a gap opening in the target sequence
!     * @param gapExtend expenses for the extension of a gap which was
started earlier.
      * @param matrix the <code>SubstitutionMatrix</code> object to use.
      */
    public SmithWaterman(double match, double replace, double insert,
double delete, double gapExtend, SubstitutionMatrix matrix)
    {
!     super(insert, delete, gapExtend, match, replace, matrix);
!     this.match      = -match;
!     this.replace    = -replace;
!     this.insert     = -insert;
!     this.delete     = -delete;
!     this.gapExt     = -gapExtend;
!     this.subMatrix  = matrix;
!     this.alignment  = "";
    }
 
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for an insert operation to the specified value.
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param ins costs for a single insert operation
-     */
-   public void setInsert(double ins) {
-     this.insert = -ins;
-   }
- 
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a delete operation to the specified value. 
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param del costs for a single deletion operation
-     */
-   public void setDelete(double del) {
-     this.delete = -del;
-   }
- 
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for an extension of any gap (insert or delete)
to the 
-     * specified value.
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1. 
-     * @param ge costs for any gap extension
-     */
-   public void setGapExt(double ge) {
-     this.gapExt = -ge;
-   }
- 
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a match operation to the specified value. 
-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param ma costs for a single match operation
-     */
-   public void setMatch(double ma) {
-     this.match = -ma;
-   }
- 
-   /** Overrides the method inherited from the NeedlemanWunsch and 
-     * sets the penalty for a replace operation to the specified value.

-     * Reason: internaly scores are used instead of penalties so that 
-     * the value is muliplied with -1.
-     * @param rep costs for a single replace operation
-     */
-   public void setReplace(double rep) {
-     this.replace = -rep;
-   } 
- 
 
    /** Overrides the method inherited from the NeedlemanWunsch and
performs only a local alignment.
      * It finds only the longest common subsequence. This is good for
the beginning, but it might
--- 53,79 ----
  public class SmithWaterman extends NeedlemanWunsch
  {
 
    private double[][] scoreMatrix; 
 
 
    /** Constructs the new SmithWaterman alignment object. Alignments
are only performed,
      * if the alphabet of the given <code>SubstitutionMatrix</code>
equals the alpabet of
!     * both the query and the target <code>Sequence</code>. If the
expenses for gap extension
      * are equal to the cost of starting a gap (delete or insert), no
affine gap penalties
      * are used, which saves memory.
      *
!     * @param match expenses for a match (usually < 0)
!     * @param replace expenses for a replace operation (usually > 0)
!     * @param insert expenses for a gap opening in the query sequence
(usually > 0)
!     * @param delete expenses for a gap opening in the target sequence
(usually > 0)
!     * @param gapExtend expenses for the extension of a gap which was
started earlier (usually >= 0)
      * @param matrix the <code>SubstitutionMatrix</code> object to use.
      */
    public SmithWaterman(double match, double replace, double insert,
double delete, double gapExtend, SubstitutionMatrix matrix)
    {
!     super(match, replace, insert, delete, gapExtend, matrix);
    }
 
 
    /** Overrides the method inherited from the NeedlemanWunsch and
performs only a local alignment.
      * It finds only the longest common subsequence. This is good for
the beginning, but it might
***************
*** 186,193 ****
          for (i=1; i<=query.length();   i++)
            for (j=1; j<=subject.length(); j++)
            {
!             E[i][j] = Math.max(E[i][j-1], scoreMatrix[i][j-1] +
insert) + gapExt;
!             F[i][j] = Math.max(F[i-1][j], scoreMatrix[i-1][j] +
delete) + gapExt;
              scoreMatrix[i][j] = max(0.0, E[i][j], F[i][j],
scoreMatrix[i-1][j-1] + matchReplace(query, subject, i, j));
 
              if (scoreMatrix[i][j] > scoreMatrix[maxI][maxJ]) {
--- 124,131 ----
          for (i=1; i<=query.length();   i++)
            for (j=1; j<=subject.length(); j++)
            {
!             E[i][j] = Math.max(E[i][j-1], scoreMatrix[i][j-1] -
insert) - gapExt;
!             F[i][j] = Math.max(F[i-1][j], scoreMatrix[i-1][j] -
delete) - gapExt;
              scoreMatrix[i][j] = max(0.0, E[i][j], F[i][j],
scoreMatrix[i-1][j-1] + matchReplace(query, subject, i, j));
 
              if (scoreMatrix[i][j] > scoreMatrix[maxI][maxJ]) {
***************
*** 223,229 ****
                // Insert || finish gap if extended gap is opened
                } else if (scoreMatrix[i][j] == E[i][j] ||
gap_extend[0]) {
                  //check if gap has been extended or freshly opened
!                 gap_extend[0] = (E[i][j] != scoreMatrix[i][j-1] +
insert + gapExt);
 
                  align[0] = '-' + align[0];
                  align[1] = st.tokenizeSymbol(subject.symbolAt(j--)) +
align[1];
--- 161,167 ----
                // Insert || finish gap if extended gap is opened
                } else if (scoreMatrix[i][j] == E[i][j] ||
gap_extend[0]) {
                  //check if gap has been extended or freshly opened
!                 gap_extend[0] = (E[i][j] != scoreMatrix[i][j-1] -
insert - gapExt);
 
                  align[0] = '-' + align[0];
                  align[1] = st.tokenizeSymbol(subject.symbolAt(j--)) +
align[1];
***************
*** 232,238 ****
                // Delete || finish gap if extended gap is opened
                } else {
                  //check if gap has been extended or freshly opened
!                 gap_extend[1] = (F[i][j] != scoreMatrix[i-1][j] +
delete + gapExt);
 
                  align[0] = st.tokenizeSymbol(query.symbolAt(i--)) +
align[0];
                  align[1] = '-' + align[1];
--- 170,176 ----
                // Delete || finish gap if extended gap is opened
                } else {
                  //check if gap has been extended or freshly opened
!                 gap_extend[1] = (F[i][j] != scoreMatrix[i-1][j] -
delete - gapExt);
 
                  align[0] = st.tokenizeSymbol(query.symbolAt(i--)) +
align[0];
                  align[1] = '-' + align[1];
***************
*** 257,264 ****
 
              scoreMatrix[i][j] = max(
                0.0,
!               scoreMatrix[i-1][j]   + delete,
!               scoreMatrix[i][j-1]   + insert,
                scoreMatrix[i-1][j-1] + matchReplace(query, subject, i,
j)
              );
 
--- 195,202 ----
 
              scoreMatrix[i][j] = max(
                0.0,
!               scoreMatrix[i-1][j]   - delete,
!               scoreMatrix[i][j-1]   - insert,
                scoreMatrix[i-1][j-1] + matchReplace(query, subject, i,
j)
              );
 
***************
*** 291,297 ****
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
 
                  // Insert
!                 } else if (scoreMatrix[i][j] == scoreMatrix[i][j-1] +
insert) {
                    align[0] = '-' + align[0];
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    path     = ' ' + path;
--- 229,235 ----
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
 
                  // Insert
!                 } else if (scoreMatrix[i][j] == scoreMatrix[i][j-1] -
insert) {
                    align[0] = '-' + align[0];
                    align[1] = st.tokenizeSymbol(subject.symbolAt(j--))
+ align[1];
                    path     = ' ' + path;
***************
*** 382,409 ****
      if ((y > z)) return y;
      return z;
    }
- 
-   /** This method computes the scores for the substution of the i-th
symbol
-    * of query by the j-th symbol of subject. 
-    * 
-    * @param query The query sequence
-    * @param subject The target sequence
-    * @param i The position of the symbol under consideration within
the 
-    *        query sequence (starting from one)
-    * @param j The position of the symbol under consideration within
the
-    *        target sequence
-    * @return The score for the given substitution.
-    */
-   private double matchReplace(Sequence query, Sequence subject, int i,
int j) {
-     try {
-       return subMatrix.getValueAt(query.symbolAt(i),
subject.symbolAt(j));
-     } catch (Exception exc) {
-       if (query.symbolAt(i).getMatches().contains(subject.symbolAt(j))
||
-         subject.symbolAt(j).getMatches().contains(query.symbolAt(i)))
-         return match;
-       return replace;
-     }
-   }
- 
- 
  }
--- 320,323 ----

_______________________________________________
biojava-dev mailing list
biojava-dev at lists.open-bio.org
http://lists.open-bio.org/mailman/listinfo/biojava-dev