Model-based prediction of sequence alignment quality

Ahola, Virpi; Aittokallio, Tero; Vihinen, Mauno; Uusipaikka, Esa

doi:10.1093/bioinformatics/btn414

Cited by 21 publications

(26 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In these experiments, we compare parameter advisors that use our estimator and five other estimators from the literature: COFFEE (Notredame et al, 1998), NorMD , MOS (Lassmann and Sonnhammer, 2005), HoT (Landan and Graur, 2007), and PredSP (Ahola et al, 2008). (In terms of our earlier categorization of estimators, COFFEE, NorMD, and PredSP are scoring-function-based, while MOS and HoT are support-based.)…”

Section: Resultsmentioning

confidence: 99%

“…These approaches follow two general strategies for estimating the accuracy with which a computed alignment recovers the unknown correct alignment. 2 The first general strategy, which we call scoring-function-based, is to develop a new scoring function on alignments that ideally is correlated with accuracy (Notredame et al (1998), Thompson et al (2001), Pei and Grishin (2001), and Ahola et al (2008)). These scoring functions combine local attributes of an alignment into a score, and typically include a measure of the conservation of amino acids in alignment columns.…”

Section: Related Workmentioning

confidence: 99%

“…Scoring-function-based approaches. Approaches that assess alignment quality via a scoring function include COFFEE (Notredame, et al, 1998), AL2CO (Pei and Grishin, 2001), NorMD , and PredSP (Ahola et al, 2008). We briefly describe each in turn.…”

Section: Related Workmentioning

confidence: 99%

“…PredSP (Ahola et al, 2008) fits a beta distribution from statistics to the true accuracies associated with a sample of alignments, where the mean and variance of the distribution are transforms of a linear…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

Accuracy Estimation and Parameter Advising for Protein Multiple Sequence Alignment

Kececioglu

DeBlasio

2013

Journal of Computational Biology

View full text Add to dashboard Cite

We develop a novel and general approach to estimating the accuracy of multiple sequence alignments without knowledge of a reference alignment, and use our approach to address a new task that we call parameter advising: the problem of choosing values for alignment scoring function parameters from a given set of choices to maximize the accuracy of a computed alignment.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. Compared to prior approaches for estimating accuracy, our new approach (a) introduces novel feature functions that measure nonlocal properties of an alignment yet are fast to evaluate, (b) considers more general classes of estimators beyond linear combinations of features, and (c) develops new regression formulations for learning an estimator from examples; in addition, for parameter advising, we (d) determine the optimal parameter set of a given cardinality, which specifies the best parameter values from which to choose.Our estimator, which we call Facet (for ''feature-based accuracy estimator''), yields a parameter advisor that on the hardest benchmarks provides more than a 27% improvement in accuracy over the best default parameter choice, and for parameter advising significantly outperforms the best prior approaches to assessing alignment quality.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Accuracy Estimation and Parameter Advising for Protein Multiple Sequence Alignment

Kececioglu

DeBlasio

2013

Journal of Computational Biology

View full text Add to dashboard Cite

show abstract

“…For boundedness to hold when using fixed feature weights, the feature functions themselves must also be bounded. Hence we assume that the feature functions also have the range [0,1]. We can then guarantee that the estimator has range [0, 1] by ensuring that coefficients found by the training process yield a convex combination of features.…”

Section: Introductionmentioning

confidence: 99%

Estimating the Accuracy of Multiple Alignments and its Use in Parameter Advising

DeBlasio

Wheeler

Kececioglu

2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

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.

show abstract

Planning the Human Variome Project: The Spain report

Kaput¹,

Cotton²,

Hardman³

et al. 2009

Hum. Mutat.

Self Cite

View full text Add to dashboard Cite

The remarkable progress in characterizing the human genome sequence, exemplified by the Human Genome Project and the HapMap Consortium, has led to the perception that knowledge and the tools (e.g., microarrays) are sufficient for many if not most biomedical research efforts. A large amount of data from diverse studies proves this perception inaccurate at best, and at worst, an impediment for further efforts to characterize the variation in the human genome. Since variation in genotype and environment are the fundamental basis to understand phenotypic variability and heritability at the population level, identifying the range of human genetic variation is crucial to the development of personalized nutrition and medicine. The Human Variome Project (HVP; http://www.humanvariomeproject.org/) was proposed initially to systematically collect mutations that cause human disease and create a cyber infrastructure to link locus specific databases (LSDB). We report here the discussions and recommendations from the 2008 HVP planning meeting held in San Feliu de Guixols, Spain, in May 2008.

show abstract

Model-based prediction of sequence alignment quality

Abstract: The method is freely available at http://www.mtt.fi/AlignmentQuality/.

Cited by 21 publications

References 37 publications

Accuracy Estimation and Parameter Advising for Protein Multiple Sequence Alignment

Accuracy Estimation and Parameter Advising for Protein Multiple Sequence Alignment

Estimating the Accuracy of Multiple Alignments and its Use in Parameter Advising

Planning the Human Variome Project: The Spain report

Contact Info

Product

Resources

About