Jove
Visualize
Contact Us
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

Related Experiment Videos

Empirical statistical estimates for sequence similarity searches

W R Pearson1

  • 1Department of Biochemistry, University of Virginia, Charlottesville 22908, USA.

Journal of Molecular Biology
|March 26, 1998
PubMed
Summary
This summary is machine-generated.

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

Empirical determination of effective gap penalties for sequence comparison.

Bioinformatics (Oxford, England)·2002
Same author

Training for bioinformatics and computational biology.

Bioinformatics (Oxford, England)·2001
Same author

Flexible sequence similarity searching with the FASTA3 program package.

Methods in molecular biology (Clifton, N.J.)·1999
Same author

Evolution of protein sequences and structures.

Journal of molecular biology·1999
Same author

Generalized neighbor-joining: more reliable phylogenetic tree reconstruction.

Molecular biology and evolution·1999
Same author

Panning for genes--A visual strategy for identifying novel gene orthologs and paralogs.

Genome research·1999

The FASTA and Smith-Waterman algorithms now provide accurate statistical estimates for sequence similarity scores, even with gaps. Length-corrected scores significantly improve the sensitivity of database searches for inferring sequence homology.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Sequence comparison is crucial for understanding protein and DNA function.
  • Accurate statistical estimation of local sequence similarity scores, especially with gaps, has been a challenge.
  • Existing methods may not fully account for library sequence length effects.

Purpose of the Study:

  • To modify the FASTA package for accurate statistical estimates of local sequence similarity scores with gaps.
  • To evaluate different methods for correcting similarity scores for library sequence length.
  • To improve the sensitivity and accuracy of sequence database searches.

Main Methods:

  • Utilized extreme value distribution based on mean and variance of scores from unrelated sequences.

Related Experiment Videos

  • Applied length correction methods, including regression-scaling and Altschul-Gish scaling.
  • Evaluated performance using protein families from PIR39 and Prosite/SwissProt databases.
  • Compared FASTA and Smith-Waterman scores for protein/protein, DNA/DNA, and protein/translated-DNA comparisons.

    • Main Results:

    • Modified FASTA provides accurate statistical estimates for similarity scores with gaps.
    • Length-corrected similarity scores (regression-scaled and Altschul-Gish scaled) significantly outperform unscaled scores.
    • Regression-scaled scores performed slightly better on Prosite/SwissProt, while Altschul-Gish scaled scores performed best on the PIR database.
    • Probability estimates derived from score distributions are generally conservative.

    Conclusions:

    • Length-corrected similarity scores enhance the sensitivity of sequence database searches.
    • Statistical parameters derived from score distributions accurately estimate significance for inferring sequence homology.
    • The modified FASTA package offers improved statistical rigor for sequence comparison.