Re: [Seqan-dev] scoring schemes

[Samuel Flores <samuelfloresc@gmail.com>]

Thanks for your help! However the alignment is not right.  The second sequence lines up about 100 residues from where it should, whereas the "Simple" scoring scheme works reasonably.  See the "LMIS", which should line up at residue 251 on the first sequence.  I'm using -1 for both gap and gapOpen:

    200     .    :    .    :    .    :    .    :    .    : 

        NVAHPASSTKVDKKIEPRGPTIKPCPPCKCPAPNLLGGPSVFIFPPKIKD

                                                          

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

    250     .    :    .    :    .    :    .    :    .    : 

        VLMISLSPIVTCVVVDVSEDDPDVQISWFVNNVEVHTAQTQTHREDYNST

                                                          

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

    300     .    :    .    :    .    :    .    :    .    : 

        LRVVSALPIQHQDWMSGKEFKCKVNNKDLPAPIERTISKPKGSVRAPQVY

                                               |      |   

        ---------------------------------PELLGGPSVFLFPPKPK

    350     .    :    .    :    .    :    .    :    .    : 

        VLPPPEEEMTKKQVTLTCMVTDFMPEDIYVEWTNNGKTELNYKNTEPVLD

                     |       |                       |    

        DTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNS

    400     .    :    .    :    .    :    .    :     

        SDGSYFMYSKLRVEKKNWVERNSYSCSVVHEGLHNHHTTKSFSR

                  |     |  |                   |    

        TYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQ

Here's the code snippet:

1002             seqan::Score<int, seqan::ScoreMatrix<seqan::AminoAcid, seqan::Pam120_> > scorePam120(gap,gapOpenPenalty);

1003             int score = globalAlignment(align, scorePam120 );

Maybe I need to specify the algorithm as well?

Also, how can I retrieve the aligned residue numbers?  Right now I'm just retrieving the two aligned sequences and assuming they are aligned if neither character is a '-': 

1022                 if ((String(seqan::row (align,0)[i]).compare("-")  != 0  )  &&

1023                     (String(seqan::row (align,1)[i]).compare("-")  != 0  )) { /* aligned residues*/ }

This works well enough for very simple, high sequence identity alignments.  However I'd really like to know which residues are positively aligned ( "|" in the above alignment).

Thanks

Sam

Hey Sam, 

if you want to use scoring matrices you need to use a different specialization of our Score class. To use scoring schemes like the Pam or Blosum matrices you need to use the following code: 

Score<int, ScoreMatrix<AminoAcid, Pam120_> > scorePam120(gap, gapOpen);

You can also use our shortcuts to make the code look nicer. You can find a list of all available shortcuts here: http://docs.seqan.de/seqan/1.4.1/SPEC_Score+_Matrix.html.

For the example above this would look like:

Pam120 scorePam120(gap, gapOpen);

I hope that will solve your problem!

Kind regards,

René Rahn

On Feb 17, 2014, at 12:32 AM, Samuel Flores <samuelfloresc@gmail.com> wrote:

Guys,

I can't seem to get globalAlignment to use any scoring scheme otehr than seqan::Simple:

           TAlign align;
           seqan::resize(rows(align), 2);
           assignSource(row(align,0),seqA);
           assignSource(row(align,1),seqB);
           int score = globalAlignment(align, seqan::Score<int,seqan::Simple>(0,-1,-1,gapOpenPenalty));

I tried changing the last line to just:

           int score = globalAlignment(align, seqan::Score<int,Pam>(0,-1,-1,gapOpenPenalty));

and every conceivable variant thereof.  None seem to be recognized, e.g.:


/Users/Sam/svn/RNAToolbox/trunk/src/ParameterReader.cpp:1000:65: error: use of undeclared identifier 'Pam'


Can someone tell me how to switch to Blossum, PAM, or some other?

Thanks

Sam
_______________________________________________
seqan-dev mailing list
seqan-dev@lists.fu-berlin.de
https://lists.fu-berlin.de/listinfo/seqan-dev

---

René Rahn

Ph.D. Student

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

Mail:  rene.rahn@fu-berlin.de

Phone: +49 (0)30 838 75 277

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

Algorithmic Bioinformatics (ABI)

Department of Computer Science 

Room  018

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

Freie Universität Berlin

Takustraße 9 

14195 Berlin

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

_______________________________________________
seqan-dev mailing list
seqan-dev@lists.fu-berlin.de
https://lists.fu-berlin.de/listinfo/seqan-dev