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 Video

Updated: Apr 24, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

2.0K

FiLM-Enhanced Biologically Informed Neural Networks for Multiclass Omics Analysis and Biomarker Discovery.

Wei Liu1, Zhenxiang Zheng2, Xujie Wang3

  • 1Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Genome Analysis Laboratory, Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, P.R. China.

Analytical Chemistry
|April 22, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same authorSame journal

Can Large Language Models Translate MS/MS into Molecular Caption?

Analytical chemistry·2026
Same author

Unified Multitask Modeling for Retention Time Prediction Across Chromatographic Conditions.

Analytical chemistry·2026
Same author

ICVAE: Interpretable Conditional Variational Autoencoder for De Novo Molecular Design.

International journal of molecular sciences·2025
Same author

DerivaPredict: A User-Friendly Tool for Predicting and Evaluating Active Derivatives of Natural Products.

Molecules (Basel, Switzerland)·2025
Same author

Spike-In Proteome Enhances Data-Independent Acquisition for Thermal Proteome Profiling.

Analytical chemistry·2024
Same author

FederEI: Federated Library Matching Framework for Electron Ionization Mass Spectrum Based Compound Identification.

Analytical chemistry·2024

Biologically informed neural networks with feature-wise linear modulation (BINN-FiLM) improve multiclass proteomic data analysis. This interpretable model captures class-specific biological rewiring, outperforming standard methods for disease discrimination.

Area of Science:

  • Computational biology
  • Bioinformatics
  • Machine learning

Background:

  • High-throughput proteomics offers detailed molecular phenotyping but faces challenges in predictive modeling due to data complexity.
  • Biologically informed neural networks (BINNs) integrate pathway knowledge for interpretable omics data analysis.

Purpose of the Study:

  • To introduce BINN-FiLM, an extension of BINNs using feature-wise linear modulation for enhanced multiclass classification of proteomic data.
  • To capture class-dependent biological rewiring and adapt pathway activations to disease-specific contexts.

Main Methods:

  • Developed BINN-FiLM, integrating feature-wise linear modulation (FiLM) into a fixed Reactome pathway hierarchy.
  • Reformulated multiclass problems as multitask binary learning for adaptive pathway activations.

More Related Videos

Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
08:51

Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts

Published on: September 20, 2024

2.1K

Related Experiment Videos

Last Updated: Apr 24, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

2.0K
Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
08:51

Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts

Published on: September 20, 2024

2.1K
  • Utilized SHAP for model interpretation to identify key proteins and pathways.

    • Main Results:

    • BINN-FiLM consistently outperformed conventional BINNs and standard machine learning models on three multiclass proteomic datasets.
    • FiLM modulation parameters successfully revealed task-specific pathway activity.
    • SHAP analysis identified key proteins and pathways driving disease discrimination.

    Conclusions:

    • BINN-FiLM provides a powerful and interpretable approach for analyzing complex, high-dimensional proteomic data in multiclass settings.
    • The model's ability to capture class-specific biological rewiring enhances its effectiveness in disease-related omics studies.