COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator

<h3>Background</h3><p dir="ltr">The post-genomic era with its wealth of sequences gave rise to a broad range of protein residue-residue contact detecting methods. Although various coevolution methods such as PSICOV, DCA and plmDCA provide correct contact predictions, they...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Reda Rawi (391865) (author)
مؤلفون آخرون: Raghvendra Mall (581171) (author), Khalid Kunji (828224) (author), Mohammed El Anbari (767963) (author), Michael Aupetit (3582545) (author), Ehsan Ullah (2698921) (author), Halima Bensmail (10400) (author)
منشور في: 2016
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513556544749568
author Reda Rawi (391865)
author2 Raghvendra Mall (581171)
Khalid Kunji (828224)
Mohammed El Anbari (767963)
Michael Aupetit (3582545)
Ehsan Ullah (2698921)
Halima Bensmail (10400)
author2_role author
author
author
author
author
author
author_facet Reda Rawi (391865)
Raghvendra Mall (581171)
Khalid Kunji (828224)
Mohammed El Anbari (767963)
Michael Aupetit (3582545)
Ehsan Ullah (2698921)
Halima Bensmail (10400)
author_role author
dc.creator.none.fl_str_mv Reda Rawi (391865)
Raghvendra Mall (581171)
Khalid Kunji (828224)
Mohammed El Anbari (767963)
Michael Aupetit (3582545)
Ehsan Ullah (2698921)
Halima Bensmail (10400)
dc.date.none.fl_str_mv 2016-12-15T03:00:00Z
dc.identifier.none.fl_str_mv 10.1186/s12859-016-1400-3
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/COUSCOus_improved_protein_contact_prediction_using_an_empirical_Bayes_covariance_estimator/27094528
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biological sciences
Bioinformatics and computational biology
Residue-residue contact prediction
Shrinkage
GLasso
dc.title.none.fl_str_mv COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <h3>Background</h3><p dir="ltr">The post-genomic era with its wealth of sequences gave rise to a broad range of protein residue-residue contact detecting methods. Although various coevolution methods such as PSICOV, DCA and plmDCA provide correct contact predictions, they do not completely overlap. Hence, new approaches and improvements of existing methods are needed to motivate further development and progress in the field. We present a new contact detecting method, COUSCOus, by combining the best shrinkage approach, the empirical Bayes covariance estimator and GLasso.</p><h3>Results</h3><p dir="ltr">Using the original PSICOV benchmark dataset, COUSCOus achieves mean accuracies of 0.74, 0.62 and 0.55 for the top L/10 predicted long, medium and short range contacts, respectively. In addition, COUSCOus attains mean areas under the precision-recall curves of 0.25, 0.29 and 0.30 for long, medium and short contacts and outperforms PSICOV. We also observed that COUSCOus outperforms PSICOV w.r.t. Matthew’s correlation coefficient criterion on full list of residue contacts. Furthermore, COUSCOus achieves on average 10% more gain in prediction accuracy compared to PSICOV on an independent test set composed of CASP11 protein targets. Finally, we showed that when using a simple random forest meta-classifier, by combining contact detecting techniques and sequence derived features, PSICOV predictions should be replaced by the more accurate COUSCOus predictions.</p><h3>Conclusion</h3><p dir="ltr">We conclude that the consideration of superior covariance shrinkage approaches will boost several research fields that apply the GLasso procedure, amongst the presented one of residue-residue contact prediction as well as fields such as gene network reconstruction.</p><h2>Other Information</h2><p dir="ltr">Published in: BMC Bioinformatics<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1186/s12859-016-1400-3" target="_blank">https://dx.doi.org/10.1186/s12859-016-1400-3</a></p>
eu_rights_str_mv openAccess
id Manara2_75b6e909fa4c1917426e37149e1b7b55
identifier_str_mv 10.1186/s12859-016-1400-3
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/27094528
publishDate 2016
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimatorReda Rawi (391865)Raghvendra Mall (581171)Khalid Kunji (828224)Mohammed El Anbari (767963)Michael Aupetit (3582545)Ehsan Ullah (2698921)Halima Bensmail (10400)Biological sciencesBioinformatics and computational biologyResidue-residue contact predictionShrinkageGLasso<h3>Background</h3><p dir="ltr">The post-genomic era with its wealth of sequences gave rise to a broad range of protein residue-residue contact detecting methods. Although various coevolution methods such as PSICOV, DCA and plmDCA provide correct contact predictions, they do not completely overlap. Hence, new approaches and improvements of existing methods are needed to motivate further development and progress in the field. We present a new contact detecting method, COUSCOus, by combining the best shrinkage approach, the empirical Bayes covariance estimator and GLasso.</p><h3>Results</h3><p dir="ltr">Using the original PSICOV benchmark dataset, COUSCOus achieves mean accuracies of 0.74, 0.62 and 0.55 for the top L/10 predicted long, medium and short range contacts, respectively. In addition, COUSCOus attains mean areas under the precision-recall curves of 0.25, 0.29 and 0.30 for long, medium and short contacts and outperforms PSICOV. We also observed that COUSCOus outperforms PSICOV w.r.t. Matthew’s correlation coefficient criterion on full list of residue contacts. Furthermore, COUSCOus achieves on average 10% more gain in prediction accuracy compared to PSICOV on an independent test set composed of CASP11 protein targets. Finally, we showed that when using a simple random forest meta-classifier, by combining contact detecting techniques and sequence derived features, PSICOV predictions should be replaced by the more accurate COUSCOus predictions.</p><h3>Conclusion</h3><p dir="ltr">We conclude that the consideration of superior covariance shrinkage approaches will boost several research fields that apply the GLasso procedure, amongst the presented one of residue-residue contact prediction as well as fields such as gene network reconstruction.</p><h2>Other Information</h2><p dir="ltr">Published in: BMC Bioinformatics<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1186/s12859-016-1400-3" target="_blank">https://dx.doi.org/10.1186/s12859-016-1400-3</a></p>2016-12-15T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1186/s12859-016-1400-3https://figshare.com/articles/journal_contribution/COUSCOus_improved_protein_contact_prediction_using_an_empirical_Bayes_covariance_estimator/27094528CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/270945282016-12-15T03:00:00Z
spellingShingle COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
Reda Rawi (391865)
Biological sciences
Bioinformatics and computational biology
Residue-residue contact prediction
Shrinkage
GLasso
status_str publishedVersion
title COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
title_full COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
title_fullStr COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
title_full_unstemmed COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
title_short COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
title_sort COUSCOus: improved protein contact prediction using an empirical Bayes covariance estimator
topic Biological sciences
Bioinformatics and computational biology
Residue-residue contact prediction
Shrinkage
GLasso