[BiO BB] Re: BiO_Bulletin_Board digest, Vol 1 #540 - 8 msgs
boris.steipe at utoronto.ca
Fri Oct 10 19:49:24 EDT 2003
sravan sravan wrote:
> I am a postgraduate in Biochemistry , with an Advanced Diploma in Bioinformatics. I looking into the concepts on sequence alignments with affine gap scoring. So I would like to forward my request to all my coulegues . If possible please suggest me the flowchart/algorithm for the alignment and scoring system for the process.
Affine gap scoring simply means that you have a fixed gap-opening and linear
gap-extension penalty. I.e. a penalty p = a + bl, where a and b are parameters
that need to be determined empirically to work well with a given scoring matrix
and l is the length of an insertion or deletion (indel). Such an affine gap
penalty is justified more from computational simplicity than from biological reality.
See for example below for an empirical determination of parameters that
emphasizes that a constant gap-opening penalty is not a good concept - it should
scale with evolutionary distance.
All this has no bearing on the fundamental principles of an optimal (e.g.
Needleman-Wunsch or Smith-Waterman) or empirical (e.g. BLAST, FASTA) alignment,
only on how the algorithm handles penalties for indels. The algorithms are
described in standard bioinformatics texts.
University of Toronto
Program in Proteomics & Bioinformatics
Departments of Biochemistry & Molecular and Medical Genetics
More information about the BBB