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

Three-parameter lognormal distribution ubiquitously found in cDNA microarray data and its application to parametric

Tomokazu Konishi1

  • 1Faculty of Bioresource Sciences, Akita Prefectural University, Akita 010-0195, Japan. konishi@akita-pu.ac.jp

BMC Bioinformatics
|January 14, 2004
PubMed
Summary

A new three-parameter lognormal distribution model offers a universal standard for normalizing microarray data. This approach simplifies data analysis and knowledge integration by addressing limitations of previous methods.

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

Challenges and Future Directions in Lipoprotein Measurement for Atherosclerosis Prevention and Treatment.

International journal of molecular sciences·2025
Same author

Distinctive features of lipoprotein profiles in stroke patients.

PloS one·2023
Same author

Theanine, a Tea-Leaf-Specific Amino Acid, Alleviates Stress through Modulation of Npas4 Expression in Group-Housed Older Mice.

International journal of molecular sciences·2023
Same author

Human lipoproteins comprise at least 12 different classes that are lognormally distributed.

PloS one·2022
Same author

Mutations in SARS-CoV-2 are on the increase against the acquired immunity.

PloS one·2022
Same author

Continuous mutation of SARS-CoV-2 during migration via three routes at the beginning of the pandemic.

PeerJ·2022

Area of Science:

  • Bioinformatics
  • Genomics
  • Statistical Modeling

Background:

  • Microarray data normalization is crucial to cancel experimental variations.
  • Parametric methods require appropriate statistical distribution models, which have often failed for microarray data.
  • Current non-parametric methods lack a universal standard, hindering data analysis and knowledge integration.

Purpose of the Study:

  • To evaluate a three-parameter lognormal distribution model for microarray data normalization.
  • To determine the model's appropriateness and identify its limitations.

Main Methods:

  • Tested a three-parameter lognormal distribution model on over 300 microarray datasets.
  • Incorporated hybridization background as a model parameter.
  • Assessed model fitting and data consistency across different intensity ranges.

Related Experiment Videos

Main Results:

  • The lognormal model demonstrated a rigorous coincidence with empirical microarray data.
  • The model provided a closer fit than previous methods, comparable to Northern analysis.
  • Reproducible measurements of z-scores and ratios were achieved within a consistent intensity range, independent of signal intensity.

Conclusions:

  • The three-parameter lognormal model can serve as a universal standard for microarray data normalization.
  • Excluding data from marginal intensity ranges (very strong or weak signals) prevents misleading analytical conclusions.
  • This approach simplifies data analysis and facilitates knowledge integration across experiments.