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

In silico microdissection of microarray data from heterogeneous cell populations.

Harri Lähdesmäki1, Llya Shmulevich, Valerie Dunmire

  • 1Institute of Signal Processing, Tampere University of Technology, P.O.Box 553, 33101 Tampere, Finland. harri.lahdesmaki@tut.fi

BMC Bioinformatics
|March 16, 2005
PubMed
Summary

This study introduces a computational method to correct microarray data for sample heterogeneity, enabling accurate gene expression analysis even with mixed cell types. The approach effectively deconvolutes mixed signals to reveal pure cell expression profiles.

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

Single-Cell Analysis of Adaptive NK and CMV-Specific T Cell Dynamics after Allogeneic HSCT.

Blood advances·2026
Same author

Mapping the future of medicine through digital twins.

Frontiers in molecular medicine·2026
Same author

The role of digital twins in P4 medicine: A paradigm for modern healthcare.

NPJ digital medicine·2025
Same author

Digital twin models for predicting venetoclax and azacitidine-induced neutropenia in patients with acute myeloid leukemia.

NPJ digital medicine·2025
Same author

Incorporating hierarchical information into multiple instance learning for patient phenotype prediction with single-cell RNA-sequencing data.

Bioinformatics (Oxford, England)·2025
Same author

Comparative analysis of the antiglioblastoma activity of adenosine A<sub>2A</sub> receptor and CK1δ blockers.

European journal of pharmacology·2025

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Genomics

Background:

  • Sample heterogeneity in microarray data, common in cancer studies, introduces significant variability.
  • This variability, due to mixed cell types, hinders accurate statistical analysis and can lead to spurious gene expression findings.
  • Existing analytical approaches are insufficient for resolving these heterogeneity-induced measurement errors.

Purpose of the Study:

  • To develop a computational framework for removing the effects of sample heterogeneity from microarray data.
  • To enable the estimation of gene expression values for pure cell populations from mixed samples.
  • To address the challenge of unknown cell type proportions in heterogeneous samples.

Main Methods:

  • A computational framework for in silico microdissection of microarray data.

Related Experiment Videos

  • Estimation of pure cell expression values and mixing percentages.
  • An optimization-based method for joint estimation when mixing percentages are unknown.
  • A model selection method for determining the correct number of cell types.

    • Main Results:

    • The proposed methods successfully reconstruct sample-specific and cell type-specific expression values from heterogeneous mixtures.
    • Accurate estimation of mixing percentages for different cell types within samples was achieved.
    • The computational framework effectively removes the confounding effects of sample heterogeneity.

    Conclusions:

    • The developed computational methods provide a robust solution for analyzing heterogeneous microarray data.
    • The approach allows for the accurate identification of true biological signals by correcting for sample composition.
    • The methods are validated on controlled cDNA microarray data, demonstrating their practical utility in biological research.