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 Concept Videos

Problem Solving: Dimensional Analysis01:08

Problem Solving: Dimensional Analysis

4.1K
Every mathematical equation that connects separate distinct physical quantities must be dimensionally consistent, which implies it must abide by two rules. For this reason, the concept of dimension is crucial. The first rule is that an equation's expressions on either side of an equality must have the exact same dimension, i.e., quantities of the same dimension can be added or removed. The second rule stipulates that all popular mathematical functions, such as exponential, logarithmic, and...
4.1K
Dimensional Analysis01:27

Dimensional Analysis

392
Dimensional analysis is a valuable technique in fluid mechanics for simplifying complex problems by reducing them into dimensionless groups. These groups capture the essential relationships between the variables involved, allowing researchers and engineers to analyze fluid flow without dealing with each variable individually. This approach reduces the number of independent variables, allowing for easier analysis and better understanding of physical phenomena.
In fluid mechanics, dimensional...
392
Downsampling01:20

Downsampling

244
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
244
Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

129
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
129
Collisions in Multiple Dimensions: Problem Solving01:06

Collisions in Multiple Dimensions: Problem Solving

4.3K
In multiple dimensions, the conservation of momentum applies in each direction independently. Hence, to solve collisions in multiple dimensions, we should write down the momentum conservation in each direction separately. To help understand collisions in multiple dimensions, consider an example.
A small car of mass 1,200 kg traveling east at 60 km/h collides at an intersection with a truck of mass 3,000 kg traveling due north at 40 km/h. The two vehicles are locked together. What is the...
4.3K
Reducing Line Loss01:18

Reducing Line Loss

190
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
190

You might also read

Related Articles

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

Sort by
Same journal

Popformer: Learning general signatures of positive selection with a self-supervised transformer.

PLoS computational biology·2026
Same journal

Representational capacity shapes both the "whether" and "how" of social learning.

PLoS computational biology·2026
Same journal

Ten quick tips to SNIFF out sustainable and secure scientific software.

PLoS computational biology·2026
Same journal

Early reduction in aversive Pavlovian bias as a mediator of anhedonia improvement during Behavioural Activation in realistic treatment settings.

PLoS computational biology·2026
Same journal

TB-SERS analyzer: Analysis tool for tuberculosis prediction based on Raman spectroscopy with machine learning and convolutional neural network.

PLoS computational biology·2026
Same journal

Application of a uniaxial force by pulling the skin around the mammary gland may affect the orientation of the ducts and the length of the mammary ductal network: Findings from computational modeling and laboratory experiments.

PLoS computational biology·2026

Related Experiment Video

Updated: Sep 4, 2025

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

1.2K

Haisu: Hierarchically supervised nonlinear dimensionality reduction.

Kevin Christopher VanHorn1, Murat Can Çobanoğlu1

  • 1Lyda Hill Department of Bioinformatics, University of Texas Southwestern Medical Center, Dallas, Texas, United States of America.

Plos Computational Biology
|July 21, 2022
PubMed
Summary

We introduce Haisu, a novel method to integrate hierarchical labels into nonlinear dimensionality reduction. This approach enhances pattern discovery in complex datasets, including single-cell RNA sequencing data, by respecting known data structures.

More Related Videos

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

7.0K
A Psychophysics Paradigm for the Collection and Analysis of Similarity Judgments
08:12

A Psychophysics Paradigm for the Collection and Analysis of Similarity Judgments

Published on: March 1, 2022

2.6K

Related Experiment Videos

Last Updated: Sep 4, 2025

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

1.2K
Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations
12:27

Large-scale Reconstructions and Independent, Unbiased Clustering Based on Morphological Metrics to Classify Neurons in Selective Populations

Published on: February 15, 2017

7.0K
A Psychophysics Paradigm for the Collection and Analysis of Similarity Judgments
08:12

A Psychophysics Paradigm for the Collection and Analysis of Similarity Judgments

Published on: March 1, 2022

2.6K

Area of Science:

  • Computational biology
  • Data visualization
  • Bioinformatics

Background:

  • Nonlinear dimensionality reduction (NLDR) techniques like t-SNE, UMAP, and PHATE are crucial for visualizing high-dimensional biological data.
  • Incorporating prior biological knowledge, such as class labels and their hierarchies, can significantly improve the interpretability of these visualizations.
  • Existing NLDR methods often lack direct mechanisms to integrate structured label information, limiting their utility in complex biological contexts.

Purpose of the Study:

  • To develop a novel strategy, termed Haisu, for incorporating hierarchical supervised label information into existing NLDR techniques.
  • To enhance the discovery and understanding of underlying biological patterns, particularly those influenced by parent-child relationships.
  • To demonstrate the utility of Haisu in both fully supervised and semi-supervised learning scenarios.

Main Methods:

  • Extension of t-SNE, UMAP, and PHATE to incorporate known or predicted class labels.
  • Mathematical perturbation of the high-dimensional space to guide the manifold towards respecting label hierarchies.
  • Application of the Haisu strategy across multiple single-cell RNA sequencing (scRNA-seq) datasets.

Main Results:

  • Demonstrated efficacy of Haisu on various scRNA-seq datasets, showcasing improved visualization interpretability.
  • Validation of Haisu's capability to aid pattern discovery in datasets with hierarchical structures.
  • Successful application in semi-supervised settings, where only a subset of data points are labeled.

Conclusions:

  • Haisu provides a flexible and effective method for integrating hierarchical label information into popular NLDR techniques.
  • The approach preserves the core characteristics of the original visualization methods while enhancing them with label-guided structure.
  • Haisu offers a valuable tool for biological data analysis, facilitating deeper insights into complex datasets and relationships.