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

Inferring latent task structure for Multitask Learning by Multiple Kernel Learning.

Christian Widmer1, Nora C Toussaint, Yasemin Altun

  • 1Friedrich Miescher Laboratory, Max Planck Society, Spemannstr, 39, 72076 Tübingen, Germany. cwidmer@tuebingen.mpg.de

BMC Bioinformatics
|November 2, 2010
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Brain Abscess l: Introduction01:26

Brain Abscess l: Introduction

A brain abscess is a focal, intracerebral infection characterized by a localized collection of pus within the brain parenchyma, resulting from microbial invasion and the body’s inflammatory response. It progresses through stages: early and late cerebritis, followed by early and late capsule formation, reflecting tissue destruction, immune response, and eventual encapsulation.Etiology and PathogenesisCausative organisms vary with source and host factors, often involving polymicrobial infections,...

You might also read

Related Articles

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

Sort by
Same author

Representation learning for multi-modal spatially resolved transcriptomics data.

Bioinformatics (Oxford, England)·2026
Same author

Estimation of Physiological Metrics from Resting ECGs Using Deep Learning in the UK Biobank, Including submaximal exercise derived V̇O <sub>2</sub> max, Body Fat Percentage, and Grip Strength.

medRxiv : the preprint server for health sciences·2026
Same author

RMS: a ML-based system for ICU respiratory monitoring and resource planning.

NPJ digital medicine·2025
Same author

Comparative cost analysis of a diagnostic multi-omics platform for decision support in advanced cancer - results from the Tumor Profiler Melanoma project.

NPJ precision oncology·2025
Same author

ImmunoPepper: extracting personalized peptides from complex splicing graphs.

Bioinformatics (Oxford, England)·2025
Same author

Efficient and accurate search in petabase-scale sequence repositories.

Nature·2025
Same journal

OpenIMC: an open-source platform for analyzing single-cell and spatial proteomics by imaging mass cytometry.

BMC bioinformatics·2026
Same journal

NAP: an open source pipeline for cross-domain microbiome profiling using Nanopore sequencing-derived amplicon data.

BMC bioinformatics·2026
Same journal

SurvGME: an R package for survival analysis with graphical and measurement error models.

BMC bioinformatics·2026
Same journal

SimMapNet: a Bayesian framework for gene regulatory network inference using gene ontology similarities as external hint.

BMC bioinformatics·2026
Same journal

Dual channel drug-drug interactions extraction based on cross attention.

BMC bioinformatics·2026
Same journal

FeSseqdb: a curated sequence-level database and interpretable machine learning framework for identifying iron-sulfur proteins.

BMC bioinformatics·2026
See all related articles

This study introduces a novel Multitask Learning (MTL) approach that simultaneously learns task similarities and classifiers. The method, applied to Computational Biology, improves performance by refining or discovering relationships between different biological datasets.

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Machine Learning

Background:

  • Limited training data hinders Machine Learning (ML) in Computational Biology.
  • Multitask Learning (MTL) leverages related problem domains to enhance model training.
  • Integrating data from multiple organisms in Bioinformatics presents challenges in managing task similarity.

Purpose of the Study:

  • To develop a novel MTL approach that learns task similarity concurrently with classifier training.
  • To address the challenge of determining task relatedness in cross-organism data integration.
  • To provide a flexible framework adaptable to prior knowledge or discovery of task relationships.

Main Methods:

  • Formulation of the MTL problem as Multiple Kernel Learning (MKL).

Related Experiment Videos

  • Utilizing the q-Norm MKL algorithm for simultaneous learning of task similarities and classifiers.
  • Application to splice site and MHC-I datasets in Computational Biology.

    • Main Results:

    • Improved performance on a splice site dataset by refining hierarchical task structures.
    • Successful *ab initio* learning of task similarities for an MHC-I dataset without prior knowledge.
    • Outperformance of baseline methods in both demonstrated applications.

    Conclusions:

    • A novel MTL framework is presented that learns task similarity alongside classifiers.
    • The approach is versatile, accommodating prior knowledge or discovering task relationships.
    • Promising results achieved in Computational Biology applications, demonstrating the method's efficacy.