File: Scoring.dddoc

package info (click to toggle)
seqan 1.4.1%2Bdfsg-2
  • links: PTS, VCS
  • area: main
  • in suites: jessie, jessie-kfreebsd
  • size: 257,080 kB
  • ctags: 38,576
  • sloc: cpp: 301,711; python: 26,086; sh: 659; xml: 188; awk: 129; makefile: 53
file content (35 lines) | stat: -rw-r--r-- 1,957 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
.Page.Scoring:
..XXXcat:Tutorials
..summary:Scoring schemes used for alignments and approximative search.

.Page.Scoring.description:
..contents

..text:
Scoring schemes define the score for aligning two characters of a given alphabet and the score for gaps within @glos:alignments@.
Given an alignment between two sequences and a scoring scheme, 
    the score of the alignment can be computed as the sum of the scores for aligned character pairs
    plus the sum of the scores for all gaps.
..text:
An example for a scoring scheme is the levenshtein distance, for which each mismatch between two aligned characters costs 1 and
each character that is aligned with a gap costs 1. 
Translated into scores instead of costs, misalignments get a score of -1 and gaps a score of -1 per character.
This scoring scheme is the default for @Spec.Simple Score@.

..section:# Character Pair Scores
..text:SeqAn offers two kinds of scoring scheme:
..table:@Spec.Simple Score@|This scoring scheme differentiates only between "match" (the two aligned characters are the same) and "mismatch" (the two aligned characters are different).
The functions @Function.scoreMatch@ and @Function.scoreMismatch@ access these two values.
..table:Scoring matrices|These scoring schemes store a score value for each pair of characters. 
This value can be accessed using @Function.score@.
Examples for this kind of scoring scheme are @Shortcut.Pam120@ and @Shortcut.Blosum62@. 
The class @Spec.Score Matrix@ can be used to store arbitrary scoring matrices.

..section:# Gap Scores
..text:SeqAn scoring schemes support @glos:affine gap costs@: The score for a gap of length $n$ is $gap_open + (n-1)*gap_extend$
while $gap_open$ and $gap_extend$ are typical negative values that represent the costs for "opening a gap" and "extendig the gap".
If $gap_open == gap_entend$ (which is the default), the scoring scheme uses @glos:linear gap costs@.


..see:Page.Alignments
..see:Page.Searching