### Abstract

For as long as biologists have been computing alignments of sequences, the question of what values to use for scoring substitutions and gaps has persisted. While some choices for substitution scores are now common, largely due to convention, there is no standard for choosing gap penalties. An objective way to resolve this question is to learn the appropriate values by solving the Inverse String Alignment Problem: given examples of correct alignments, find parameter values that make the examples be optimal-scoring alignments of their strings. We present a new polynomial-time algorithm for Inverse String Alignment that is simple to implement, fast in practice, and for the first time can learn hundreds of parameters simultaneously. The approach is also flexible: minor modifications allow us to solve inverse unique alignment (find parameter values that make the examples be the unique optimal alignments of their strings), and inverse near-optimal alignment (find parameter values that make the example alignments be as close to optimal as possible). Computational results with an implementation for global alignment show that, for the first time, we can find best-possible values for all 212 parameters of the standard protein-sequence scoring-model from hundreds of alignments in a few minutes of computation.

Original language | English (US) |
---|---|

Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |

Pages | 441-455 |

Number of pages | 15 |

Volume | 3909 LNBI |

DOIs | |

State | Published - 2006 |

Event | 10th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2006 - Venice, Italy Duration: Apr 2 2006 → Apr 5 2006 |

### Publication series

Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|

Volume | 3909 LNBI |

ISSN (Print) | 03029743 |

ISSN (Electronic) | 16113349 |

### Other

Other | 10th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2006 |
---|---|

Country | Italy |

City | Venice |

Period | 4/2/06 → 4/5/06 |

### Fingerprint

### Keywords

- Affine gap penalties
- Cutting plane algorithms
- Linear programming
- Parametric sequence alignment
- Sequence analysis
- Substitution score matrices
- Supervised learning

### ASJC Scopus subject areas

- Computer Science(all)
- Biochemistry, Genetics and Molecular Biology(all)
- Theoretical Computer Science

### Cite this

*Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)*(Vol. 3909 LNBI, pp. 441-455). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3909 LNBI). https://doi.org/10.1007/11732990_37

**Simple and fast inverse alignment.** / Kececioglu, John D; Kim, Eagu.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics).*vol. 3909 LNBI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3909 LNBI, pp. 441-455, 10th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2006, Venice, Italy, 4/2/06. https://doi.org/10.1007/11732990_37

}

TY - GEN

T1 - Simple and fast inverse alignment

AU - Kececioglu, John D

AU - Kim, Eagu

PY - 2006

Y1 - 2006

N2 - For as long as biologists have been computing alignments of sequences, the question of what values to use for scoring substitutions and gaps has persisted. While some choices for substitution scores are now common, largely due to convention, there is no standard for choosing gap penalties. An objective way to resolve this question is to learn the appropriate values by solving the Inverse String Alignment Problem: given examples of correct alignments, find parameter values that make the examples be optimal-scoring alignments of their strings. We present a new polynomial-time algorithm for Inverse String Alignment that is simple to implement, fast in practice, and for the first time can learn hundreds of parameters simultaneously. The approach is also flexible: minor modifications allow us to solve inverse unique alignment (find parameter values that make the examples be the unique optimal alignments of their strings), and inverse near-optimal alignment (find parameter values that make the example alignments be as close to optimal as possible). Computational results with an implementation for global alignment show that, for the first time, we can find best-possible values for all 212 parameters of the standard protein-sequence scoring-model from hundreds of alignments in a few minutes of computation.

AB - For as long as biologists have been computing alignments of sequences, the question of what values to use for scoring substitutions and gaps has persisted. While some choices for substitution scores are now common, largely due to convention, there is no standard for choosing gap penalties. An objective way to resolve this question is to learn the appropriate values by solving the Inverse String Alignment Problem: given examples of correct alignments, find parameter values that make the examples be optimal-scoring alignments of their strings. We present a new polynomial-time algorithm for Inverse String Alignment that is simple to implement, fast in practice, and for the first time can learn hundreds of parameters simultaneously. The approach is also flexible: minor modifications allow us to solve inverse unique alignment (find parameter values that make the examples be the unique optimal alignments of their strings), and inverse near-optimal alignment (find parameter values that make the example alignments be as close to optimal as possible). Computational results with an implementation for global alignment show that, for the first time, we can find best-possible values for all 212 parameters of the standard protein-sequence scoring-model from hundreds of alignments in a few minutes of computation.

KW - Affine gap penalties

KW - Cutting plane algorithms

KW - Linear programming

KW - Parametric sequence alignment

KW - Sequence analysis

KW - Substitution score matrices

KW - Supervised learning

UR - http://www.scopus.com/inward/record.url?scp=33745804251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745804251&partnerID=8YFLogxK

U2 - 10.1007/11732990_37

DO - 10.1007/11732990_37

M3 - Conference contribution

SN - 3540332952

SN - 9783540332954

VL - 3909 LNBI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 441

EP - 455

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -