Estimating the accuracy of multiple alignments and its use in parameter advising

Dan F. Deblasio, Travis J. Wheeler, John D Kececioglu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

We develop a novel and general approach to estimating the accuracy of protein multiple sequence alignments without knowledge of a reference alignment, and use our approach to address a new problem that we call parameter advising. For protein alignments, we consider twelve independent features that contribute to a quality alignment. An accuracy estimator is learned that is a polynomial function of these features; its coefficients are determined by minimizing its error with respect to true accuracy using mathematical optimization. We evaluate this approach by applying it to the task of parameter advising: the problem of choosing alignment scoring parameters from a collection of parameter values to maximize the accuracy of a computed alignment. Our estimator, which we call Facet (for "feature-based accuracy estimator"), yields a parameter advisor that on the hardest benchmarks provides more than a 20% improvement in accuracy over the best default parameter choice, and outperforms the best prior approaches to selecting good alignments for parameter advising.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages45-59
Number of pages15
Volume7262 LNBI
DOIs
StatePublished - 2012
Event16th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2012 - Barcelona, Spain
Duration: Apr 21 2012Apr 24 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7262 LNBI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other16th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2012
CountrySpain
CityBarcelona
Period4/21/124/24/12

Fingerprint

Alignment
Estimator
Proteins
Protein
Multiple Sequence Alignment
Polynomial function
Scoring
Maximise
Polynomials
Benchmark
Optimization
Evaluate
Coefficient

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Deblasio, D. F., Wheeler, T. J., & Kececioglu, J. D. (2012). Estimating the accuracy of multiple alignments and its use in parameter advising. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7262 LNBI, pp. 45-59). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7262 LNBI). https://doi.org/10.1007/978-3-642-29627-7_5

Estimating the accuracy of multiple alignments and its use in parameter advising. / Deblasio, Dan F.; Wheeler, Travis J.; Kececioglu, John D.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7262 LNBI 2012. p. 45-59 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7262 LNBI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Deblasio, DF, Wheeler, TJ & Kececioglu, JD 2012, Estimating the accuracy of multiple alignments and its use in parameter advising. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7262 LNBI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7262 LNBI, pp. 45-59, 16th Annual International Conference on Research in Computational Molecular Biology, RECOMB 2012, Barcelona, Spain, 4/21/12. https://doi.org/10.1007/978-3-642-29627-7_5
Deblasio DF, Wheeler TJ, Kececioglu JD. Estimating the accuracy of multiple alignments and its use in parameter advising. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7262 LNBI. 2012. p. 45-59. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-29627-7_5
Deblasio, Dan F. ; Wheeler, Travis J. ; Kececioglu, John D. / Estimating the accuracy of multiple alignments and its use in parameter advising. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7262 LNBI 2012. pp. 45-59 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{cfa2f6ceced24156a8231e05758a706e,
title = "Estimating the accuracy of multiple alignments and its use in parameter advising",
abstract = "We develop a novel and general approach to estimating the accuracy of protein multiple sequence alignments without knowledge of a reference alignment, and use our approach to address a new problem that we call parameter advising. For protein alignments, we consider twelve independent features that contribute to a quality alignment. An accuracy estimator is learned that is a polynomial function of these features; its coefficients are determined by minimizing its error with respect to true accuracy using mathematical optimization. We evaluate this approach by applying it to the task of parameter advising: the problem of choosing alignment scoring parameters from a collection of parameter values to maximize the accuracy of a computed alignment. Our estimator, which we call Facet (for {"}feature-based accuracy estimator{"}), yields a parameter advisor that on the hardest benchmarks provides more than a 20{\%} improvement in accuracy over the best default parameter choice, and outperforms the best prior approaches to selecting good alignments for parameter advising.",
author = "Deblasio, {Dan F.} and Wheeler, {Travis J.} and Kececioglu, {John D}",
year = "2012",
doi = "10.1007/978-3-642-29627-7_5",
language = "English (US)",
isbn = "9783642296260",
volume = "7262 LNBI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "45--59",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Estimating the accuracy of multiple alignments and its use in parameter advising

AU - Deblasio, Dan F.

AU - Wheeler, Travis J.

AU - Kececioglu, John D

PY - 2012

Y1 - 2012

N2 - We develop a novel and general approach to estimating the accuracy of protein multiple sequence alignments without knowledge of a reference alignment, and use our approach to address a new problem that we call parameter advising. For protein alignments, we consider twelve independent features that contribute to a quality alignment. An accuracy estimator is learned that is a polynomial function of these features; its coefficients are determined by minimizing its error with respect to true accuracy using mathematical optimization. We evaluate this approach by applying it to the task of parameter advising: the problem of choosing alignment scoring parameters from a collection of parameter values to maximize the accuracy of a computed alignment. Our estimator, which we call Facet (for "feature-based accuracy estimator"), yields a parameter advisor that on the hardest benchmarks provides more than a 20% improvement in accuracy over the best default parameter choice, and outperforms the best prior approaches to selecting good alignments for parameter advising.

AB - We develop a novel and general approach to estimating the accuracy of protein multiple sequence alignments without knowledge of a reference alignment, and use our approach to address a new problem that we call parameter advising. For protein alignments, we consider twelve independent features that contribute to a quality alignment. An accuracy estimator is learned that is a polynomial function of these features; its coefficients are determined by minimizing its error with respect to true accuracy using mathematical optimization. We evaluate this approach by applying it to the task of parameter advising: the problem of choosing alignment scoring parameters from a collection of parameter values to maximize the accuracy of a computed alignment. Our estimator, which we call Facet (for "feature-based accuracy estimator"), yields a parameter advisor that on the hardest benchmarks provides more than a 20% improvement in accuracy over the best default parameter choice, and outperforms the best prior approaches to selecting good alignments for parameter advising.

UR - http://www.scopus.com/inward/record.url?scp=84860793205&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860793205&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-29627-7_5

DO - 10.1007/978-3-642-29627-7_5

M3 - Conference contribution

SN - 9783642296260

VL - 7262 LNBI

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

SP - 45

EP - 59

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

ER -