Jove
Visualize
Contact Us

Related Experiment Videos

The correlation error and finite-size correction in an ungapped sequence alignment.

Yonil Park1, John L Spouge

  • 1National Center for Biotechnology Information, National Library of Medicine, Bethesda, MD 20894, USA.

Bioinformatics (Oxford, England)
|September 10, 2002
PubMed
Summary

The BLAST p-value calculation neglects sequence correlations, but the independent diagonals model remains within a factor of 2 of true values. Finite-size corrections improve approximations for most sequence lengths.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Hybrid analysis with phylogeny and population modeling to estimate the recent founding date of a population: A case study in the origins of COVID-19 illustrates how a branching process approximation can simplify a hybrid analysis.

Mathematical biosciences·2025
Same author

Increasing Representativeness in the <i>All of Us</i> Cohort Using Inverse Probability Weighting.

medRxiv : the preprint server for health sciences·2025
Same author

The Statistics of Parametrized Syncmers in a Simple Mutation Process Without Spurious Matches.

Journal of computational biology : a journal of computational molecular cell biology·2024
Same author

Landscape of racial and ethnic health disparities in the All of Us Research Program.

Database : the journal of biological databases and curation·2024
Same author

Ancestry-attenuated effects of socioeconomic deprivation on type 2 diabetes disparities in the <i>All of Us</i> cohort.

BMC global and public health·2023
Same author

Ancestry-attenuated effects of socioeconomic deprivation on type 2 diabetes disparities in the All of Us cohort.

Research square·2023
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Sequence Alignment

Background:

  • The Basic Local Alignment Search Tool (BLAST) algorithm's random model assumes sequence independence, potentially introducing correlations in alignment matrices.
  • Analytic formulas for BLAST p-values often simplify this model, neglecting diagonal correlations and using finite-size corrections for short sequences.
  • The impact of neglected correlations versus finite-size corrections on BLAST p-values remains unclear, especially as sequence lengths increase.

Purpose of the Study:

  • To quantify the impact of neglected correlations and finite-size corrections on BLAST p-values.
  • To compare the effectiveness of different finite-size correction formulas.
  • To assess the accuracy of the independent diagonals model for BLAST p-value approximation.

Main Methods:

Related Experiment Videos

  • Computer experiments focused on ungapped sequence alignment to derive analytic formulas.
  • Analysis of the relative rates at which finite-size corrections vanish compared to neglected correlations.
  • Evaluation of the Altschul-Gish and Spouge finite-size correction methods.

Main Results:

  • The finite-size correction was found to vanish faster than neglected correlations.
  • Finite-size corrections, while underestimating BLAST p-values, significantly improved approximations for non-short sequences.
  • The Altschul-Gish correction outperformed Spouge's method.
  • The independent diagonals model provided approximations within a factor of 2 of true BLAST p-values.

Conclusions:

  • Neglected correlations in BLAST p-value calculations are practically less significant than the finite-size correction for most sequence comparisons.
  • The Altschul-Gish correction offers a superior approach for finite-size adjustments in BLAST.
  • While the independent diagonals model offers a reasonable approximation, fitting p-value parameters directly from it is not recommended.